fix: empty message render

micro-fix: remove unnecessary commit message file

.github/ISSUE_TEMPLATE/standard-bounty.yml

@@ -0,0 +1,78 @@
+name: Standard Bounty
+description: A bounty task for general framework contributions (not integration-specific)
+title: "[Bounty]: "
+labels: []
+body:
+  - type: markdown
+    attributes:
+      value: |
+        ## Standard Bounty
+
+        This issue is part of the [Bounty Program](../../docs/bounty-program/README.md).
+        **Claim this bounty** by commenting below — a maintainer will assign you within 24 hours.
+
+  - type: dropdown
+    id: bounty-size
+    attributes:
+      label: Bounty Size
+      options:
+        - "Small (10 pts)"
+        - "Medium (30 pts)"
+        - "Large (75 pts)"
+        - "Extreme (150 pts)"
+    validations:
+      required: true
+
+  - type: dropdown
+    id: difficulty
+    attributes:
+      label: Difficulty
+      options:
+        - Easy
+        - Medium
+        - Hard
+    validations:
+      required: true
+
+  - type: textarea
+    id: description
+    attributes:
+      label: Description
+      description: What needs to be done to complete this bounty.
+      placeholder: |
+        Describe the specific task, including:
+        - What the contributor needs to do
+        - Links to relevant files in the repo
+        - Any context or motivation for the change
+    validations:
+      required: true
+
+  - type: textarea
+    id: acceptance-criteria
+    attributes:
+      label: Acceptance Criteria
+      description: What "done" looks like. The PR must meet all criteria.
+      placeholder: |
+        - [ ] Criterion 1
+        - [ ] Criterion 2
+        - [ ] CI passes
+    validations:
+      required: true
+
+  - type: textarea
+    id: relevant-files
+    attributes:
+      label: Relevant Files
+      description: Links to files or directories related to this bounty.
+      placeholder: |
+        - `path/to/file.py`
+        - `path/to/directory/`
+
+  - type: textarea
+    id: resources
+    attributes:
+      label: Resources
+      description: Links to docs, issues, or external references that will help.
+      placeholder: |
+        - Related issue: #XXXX
+        - Docs: https://...

.gitignore

@@ -68,7 +68,6 @@ temp/
 exports/*
 
 .claude/settings.local.json
-.claude/skills/ship-it/
 
 .venv
 

CHANGELOG.md

@@ -1,17 +1,149 @@
 # Release Notes
 
+## v0.7.1
+
+**Release Date:** March 13, 2026
+**Tag:** v0.7.1
+
+### Chrome-Native Browser Control
+
+v0.7.1 replaces Playwright with direct Chrome DevTools Protocol (CDP) integration. The GCU now launches the user's system Chrome via `open -n` on macOS, connects over CDP, and manages browser lifecycle end-to-end -- no extra browser binary required.
+
+---
+
+### Highlights
+
+#### System Chrome via CDP
+
+The entire GCU browser stack has been rewritten:
+
+- **Chrome finder & launcher** -- New `chrome_finder.py` discovers installed Chrome and `chrome_launcher.py` manages process lifecycle with `--remote-debugging-port`
+- **Coexist with user's browser** -- `open -n` on macOS launches a separate Chrome instance so the user's tabs stay untouched
+- **Dynamic viewport sizing** -- Viewport auto-sizes to the available display area, suppressing Chrome warning bars
+- **Orphan cleanup** -- Chrome processes are killed on GCU server shutdown to prevent leaks
+- **`--no-startup-window`** -- Chrome launches headlessly by default until a page is needed
+
+#### Per-Subagent Browser Isolation
+
+Each GCU subagent gets its own Chrome user-data directory, preventing cookie/session cross-contamination:
+
+- Unique browser profiles injected per subagent
+- Profiles cleaned up after top-level GCU node execution
+- Tab origin and age metadata tracked per subagent
+
+#### Dummy Agent Testing Framework
+
+A comprehensive test suite for validating agent graph patterns without LLM calls:
+
+- 8 test modules covering echo, pipeline, branch, parallel merge, retry, feedback loop, worker, and GCU subagent patterns
+- Shared fixtures and a `run_all.py` runner for CI integration
+- Subagent lifecycle tests
+
+---
+
+### What's New
+
+#### GCU Browser
+
+- **Switch from Playwright to system Chrome via CDP** -- Direct CDP connection replaces Playwright dependency. (@bryanadenhq)
+- **Chrome finder and launcher modules** -- `chrome_finder.py` and `chrome_launcher.py` for cross-platform Chrome discovery and process management. (@bryanadenhq)
+- **Dynamic viewport sizing** -- Auto-size viewport and suppress Chrome warning bar. (@bryanadenhq)
+- **Per-subagent browser profile isolation** -- Unique user-data directories per subagent with cleanup. (@bryanadenhq)
+- **Tab origin/age metadata** -- Track which subagent opened each tab and when. (@bryanadenhq)
+- **`browser_close_all` tool** -- Bulk tab cleanup for agents managing many pages. (@bryanadenhq)
+- **Auto-track popup pages** -- Popups are automatically captured and tracked. (@bryanadenhq)
+- **Auto-snapshot from browser interactions** -- Browser interaction tools return screenshots automatically. (@bryanadenhq)
+- **Kill orphaned Chrome processes** -- GCU server shutdown cleans up lingering Chrome instances. (@bryanadenhq)
+- **`--no-startup-window` Chrome flag** -- Prevent empty window on launch. (@bryanadenhq)
+- **Launch Chrome via `open -n` on macOS** -- Coexist with the user's running browser. (@bryanadenhq)
+
+#### Framework & Runtime
+
+- **Session resume fix for new agents** -- Correctly resume sessions when a new agent is loaded. (@bryanadenhq)
+- **Queen upsert fix** -- Prevent duplicate queen entries on session restore. (@bryanadenhq)
+- **Anchor worker monitoring to queen's session ID on cold-restore** -- Worker monitors reconnect to the correct queen after restart. (@bryanadenhq)
+- **Update meta.json when loading workers** -- Worker metadata stays in sync with runtime state. (@RichardTang-Aden)
+- **Generate worker MCP file correctly** -- Fix MCP config generation for spawned workers. (@RichardTang-Aden)
+- **Share event bus so tool events are visible to parent** -- Tool execution events propagate up to parent graphs. (@bryanadenhq)
+- **Subagent activity tracking in queen status** -- Queen instructions include live subagent status. (@bryanadenhq)
+- **GCU system prompt updates** -- Auto-snapshots, batching, popup tracking, and close_all guidance. (@bryanadenhq)
+
+#### Frontend
+
+- **Loading spinner in draft panel** -- Shows spinner during planning phase instead of blank panel. (@bryanadenhq)
+- **Fix credential modal errors** -- Modal no longer eats errors; banner stays visible. (@bryanadenhq)
+- **Fix credentials_required loop** -- Stop clearing the flag on modal close to prevent infinite re-prompting. (@bryanadenhq)
+- **Fix "Add tab" dropdown overflow** -- Dropdown no longer hidden when many agents are open. (@prasoonmhwr)
+
+#### Testing
+
+- **Dummy agent test framework** -- 8 test modules (echo, pipeline, branch, parallel merge, retry, feedback loop, worker, GCU subagent) with shared fixtures and CI runner. (@bryanadenhq)
+- **Subagent lifecycle tests** -- Validate subagent spawn and completion flows. (@bryanadenhq)
+
+#### Documentation & Infrastructure
+
+- **MCP integration PRD** -- Product requirements for MCP server registry. (@TimothyZhang7)
+- **Skills registry PRD** -- Product requirements for skill registry system. (@bryanadenhq)
+- **Bounty program updates** -- Standard bounty issue template and updated contributor guide. (@bryanadenhq)
+- **Windows quickstart** -- Add default context limit for PowerShell setup. (@bryanadenhq)
+- **Remove deprecated files** -- Clean up `setup_mcp.py`, `verify_mcp.py`, `antigravity-setup.md`, and `setup-antigravity-mcp.sh`. (@bryanadenhq)
+
+---
+
+### Bug Fixes
+
+- Fix credential modal eating errors and banner staying open
+- Stop clearing `credentials_required` on modal close to prevent infinite loop
+- Share event bus so tool events are visible to parent graph
+- Use lazy %-formatting in subagent completion log to avoid f-string in logger
+- Anchor worker monitoring to queen's session ID on cold-restore
+- Update meta.json when loading workers
+- Generate worker MCP file correctly
+- Fix "Add tab" dropdown partially hidden when creating multiple agents
+
+---
+
+### Community Contributors
+
+- **Prasoon Mahawar** (@prasoonmhwr) -- Fix UI overflow on agent tab dropdown
+- **Richard Tang** (@RichardTang-Aden) -- Worker MCP generation and meta.json fixes
+
+---
+
+### Upgrading
+
+```bash
+git pull origin main
+uv sync
+```
+
+The Playwright dependency is no longer required for GCU browser operations. Chrome must be installed on the host system.
+
+---
+
+## v0.7.0
+
+**Release Date:** March 5, 2026
+**Tag:** v0.7.0
+
+Session management refactor release.
+
+---
+
+## v0.5.1
+
 **Release Date:** February 18, 2026
 **Tag:** v0.5.1
 
-## The Hive Gets a Brain
+### The Hive Gets a Brain
 
 v0.5.1 is our most ambitious release yet. Hive agents can now **build other agents** -- the new Hive Coder meta-agent writes, tests, and fixes agent packages from natural language. The runtime grows multi-graph support so one session can orchestrate multiple agents simultaneously. The TUI gets a complete overhaul with an in-app agent picker, live streaming, and seamless escalation to the Coder. And we're now provider-agnostic: Claude Code subscriptions, OpenAI-compatible endpoints, and any LiteLLM-supported model work out of the box.
 
 ---
 
-## Highlights
+### Highlights
 
-### Hive Coder -- The Agent That Builds Agents
+#### Hive Coder -- The Agent That Builds Agents
 
 A native meta-agent that lives inside the framework at `core/framework/agents/hive_coder/`. Give it a natural-language specification and it produces a complete agent package -- goal definition, node prompts, edge routing, MCP tool wiring, tests, and all boilerplate files.
 
@@ -30,7 +162,7 @@ The Coder ships with:
 - **Coder Tools MCP server** -- file I/O, fuzzy-match editing, git snapshots, and sandboxed shell execution (`tools/coder_tools_server.py`)
 - **Test generation** -- structural tests for forever-alive agents that don't hang on `runner.run()`
 
-### Multi-Graph Agent Runtime
+#### Multi-Graph Agent Runtime
 
 `AgentRuntime` now supports loading, managing, and switching between multiple agent graphs within a single session. Six new lifecycle tools give agents (and the TUI) full control:
 
@@ -44,7 +176,7 @@ await runtime.add_graph("exports/deep_research_agent")
 
 The Hive Coder uses multi-graph internally -- when you escalate from a worker agent, the Coder loads as a separate graph while the worker stays alive in the background.
 
-### TUI Revamp
+#### TUI Revamp
 
 The Terminal UI gets a ground-up rebuild with five major additions:
 
@@ -54,7 +186,7 @@ The Terminal UI gets a ground-up rebuild with five major additions:
 - **PDF attachments** -- `/attach` and `/detach` commands with native OS file dialog (macOS, Linux, Windows)
 - **Multi-graph commands** -- `/graphs`, `/graph <id>`, `/load <path>`, `/unload <id>` for managing agent graphs in-session
 
-### Provider-Agnostic LLM Support
+#### Provider-Agnostic LLM Support
 
 Hive is no longer Anthropic-only. v0.5.1 adds first-class support for:
 
@@ -66,9 +198,9 @@ The quickstart script auto-detects Claude Code subscriptions and ZAI Code instal
 
 ---
 
-## What's New
+### What's New
 
-### Architecture & Runtime
+#### Architecture & Runtime
 
 - **Hive Coder meta-agent** -- Natural-language agent builder with reference docs, guardian watchdog, and `hive code` CLI command. (@TimothyZhang7)
 - **Multi-graph agent sessions** -- `add_graph`/`remove_graph` on AgentRuntime with 6 lifecycle tools (`load_agent`, `unload_agent`, `start_agent`, `restart_agent`, `list_agents`, `get_user_presence`). (@TimothyZhang7)
@@ -79,7 +211,7 @@ The quickstart script auto-detects Claude Code subscriptions and ZAI Code instal
 - **Pre-start confirmation prompt** -- Interactive prompt before agent execution allowing credential updates or abort. (@RichardTang-Aden)
 - **Event bus multi-graph support** -- `graph_id` on events, `filter_graph` on subscriptions, `ESCALATION_REQUESTED` event type, `exclude_own_graph` filter. (@TimothyZhang7)
 
-### TUI Improvements
+#### TUI Improvements
 
 - **In-app agent picker** (Ctrl+A) -- Tabbed modal for browsing agents with metadata badges (nodes, tools, sessions, tags). (@TimothyZhang7)
 - **Runtime-optional TUI startup** -- Launches without a pre-loaded agent, shows agent picker on startup. (@TimothyZhang7)
@@ -89,7 +221,7 @@ The quickstart script auto-detects Claude Code subscriptions and ZAI Code instal
 - **Multi-graph TUI commands** -- `/graphs`, `/graph <id>`, `/load <path>`, `/unload <id>`. (@TimothyZhang7)
 - **Agent Guardian watchdog** -- Event-driven monitor that catches secondary agent failures and triggers automatic remediation, with `--no-guardian` CLI flag. (@TimothyZhang7)
 
-### New Tool Integrations
+#### New Tool Integrations
 
 | Tool                   | Description                                                                                                                                                            | Contributor        |
 | ---------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------ |
@@ -99,7 +231,7 @@ The quickstart script auto-detects Claude Code subscriptions and ZAI Code instal
 | **Google Docs**        | Document creation, reading, and editing with OAuth credential support                                                                                                  | @haliaeetusvocifer |
 | **Gmail enhancements** | Expanded mail operations for inbox management                                                                                                                          | @bryanadenhq       |
 
-### Infrastructure
+#### Infrastructure
 
 - **Default node type → `event_loop`** -- `NodeSpec.node_type` defaults to `"event_loop"` instead of `"llm_tool_use"`. (@TimothyZhang7)
 - **Default `max_node_visits` → 0 (unlimited)** -- Nodes default to unlimited visits, reducing friction for feedback loops and forever-alive agents. (@TimothyZhang7)
@@ -112,7 +244,7 @@ The quickstart script auto-detects Claude Code subscriptions and ZAI Code instal
 
 ---
 
-## Bug Fixes
+### Bug Fixes
 
 - Flush WIP accumulator outputs on cancel/failure so edge conditions see correct values on resume
 - Stall detection state preserved across resume (no more resets on checkpoint restore)
@@ -125,13 +257,13 @@ The quickstart script auto-detects Claude Code subscriptions and ZAI Code instal
 - Fix email agent version conflicts (@RichardTang-Aden)
 - Fix coder tool timeouts (120s for tests, 300s cap for commands)
 
-## Documentation
+### Documentation
 
 - Clarify installation and prevent root pip install misuse (@paarths-collab)
 
 ---
 
-## Agent Updates
+### Agent Updates
 
 - **Email Inbox Management** -- Consolidate `gmail_inbox_guardian` and `inbox_management` into a single unified agent with updated prompts and config. (@RichardTang-Aden, @bryanadenhq)
 - **Job Hunter** -- Updated node prompts, config, and agent metadata; added PDF resume selection. (@bryanadenhq)
@@ -141,7 +273,7 @@ The quickstart script auto-detects Claude Code subscriptions and ZAI Code instal
 
 ---
 
-## Breaking Changes
+### Breaking Changes
 
 - **Deprecated node types raise `RuntimeError`** -- `llm_tool_use`, `llm_generate`, `function`, `router`, `human_input` now fail instead of warning. Migrate to `event_loop`.
 - **`NodeSpec.node_type` defaults to `"event_loop"`** (was `"llm_tool_use"`)
@@ -150,7 +282,7 @@ The quickstart script auto-detects Claude Code subscriptions and ZAI Code instal
 
 ---
 
-## Community Contributors
+### Community Contributors
 
 A huge thank you to everyone who contributed to this release:
 
@@ -165,14 +297,14 @@ A huge thank you to everyone who contributed to this release:
 
 ---
 
-## Upgrading
+### Upgrading
 
 ```bash
 git pull origin main
 uv sync
 ```
 
-### Migration Guide
+#### Migration Guide
 
 If your agents use deprecated node types, update them:
 
@@ -196,12 +328,3 @@ hive code
 # Or from TUI -- press Ctrl+E to escalate
 hive tui
 ```
-
----
-
-## What's Next
-
-- **Agent-to-agent communication** -- one agent's output triggers another agent's entry point
-- **Cost visibility** -- detailed runtime log of LLM costs per node and per session
-- **Persistent webhook subscriptions** -- survive agent restarts without re-registering
-- **Remote agent deployment** -- run agents as long-lived services with HTTP APIs

Makefile

@@ -1,24 +1,31 @@
-.PHONY: lint format check test install-hooks help frontend-install frontend-dev frontend-build
+.PHONY: lint format check test test-tools test-live test-all install-hooks help frontend-install frontend-dev frontend-build
+
+# ── Ensure uv is findable in Git Bash on Windows ──────────────────────────────
+# uv installs to ~/.local/bin on Windows/Linux/macOS. Git Bash may not include
+# this in PATH by default, so we prepend it here.
+export PATH := $(HOME)/.local/bin:$(PATH)
+
+# ── Targets ───────────────────────────────────────────────────────────────────
 
 help: ## Show this help
 	@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | \
 		awk 'BEGIN {FS = ":.*?## "}; {printf "  \033[36m%-15s\033[0m %s\n", $$1, $$2}'
 
 lint: ## Run ruff linter and formatter (with auto-fix)
-	cd core && ruff check --fix .
-	cd tools && ruff check --fix .
-	cd core && ruff format .
-	cd tools && ruff format .
+	cd core && uv run ruff check --fix .
+	cd tools && uv run ruff check --fix .
+	cd core && uv run ruff format .
+	cd tools && uv run ruff format .
 
 format: ## Run ruff formatter
-	cd core && ruff format .
-	cd tools && ruff format .
+	cd core && uv run ruff format .
+	cd tools && uv run ruff format .
 
 check: ## Run all checks without modifying files (CI-safe)
-	cd core && ruff check .
-	cd tools && ruff check .
-	cd core && ruff format --check .
-	cd tools && ruff format --check .
+	cd core && uv run ruff check .
+	cd tools && uv run ruff check .
+	cd core && uv run ruff format --check .
+	cd tools && uv run ruff format --check .
 
 test: ## Run all tests (core + tools, excludes live)
 	cd core && uv run python -m pytest tests/ -v
@@ -46,4 +53,4 @@ frontend-dev: ## Start frontend dev server
 	cd core/frontend && npm run dev
 
 frontend-build: ## Build frontend for production
-	cd core/frontend && npm run build
+	cd core/frontend && npm run build

README.md

@@ -27,7 +27,7 @@
   <img src="https://img.shields.io/badge/Multi--Agent-Systems-blue?style=flat-square" alt="Multi-Agent" />
   <img src="https://img.shields.io/badge/Headless-Development-purple?style=flat-square" alt="Headless" />
   <img src="https://img.shields.io/badge/Human--in--the--Loop-orange?style=flat-square" alt="HITL" />
-  <img src="https://img.shields.io/badge/Production--Ready-red?style=flat-square" alt="Production" />
+  <img src="https://img.shields.io/badge/Browser-Use-red?style=flat-square" alt="Browser Use" />
 </p>
 <p align="center">
   <img src="https://img.shields.io/badge/OpenAI-supported-412991?style=flat-square&logo=openai" alt="OpenAI" />
@@ -37,15 +37,16 @@
 
 ## Overview
 
-Build autonomous, reliable, self-improving AI agents without hardcoding workflows. Define your goal through conversation with hive coding agent(queen), and the framework generates a node graph with dynamically created connection code. When things break, the framework captures failure data, evolves the agent through the coding agent, and redeploys. Built-in human-in-the-loop nodes, credential management, and real-time monitoring give you control without sacrificing adaptability.
+Generate a swarm of worker agents with a coding agent(queen) that control them. Define your goal through conversation with hive queen, and the framework generates a node graph with dynamically created connection code. When things break, the framework captures failure data, evolves the agent through the coding agent, and redeploys. Built-in human-in-the-loop nodes, browser use, credential management, and real-time monitoring give you control without sacrificing adaptability.
 
 Visit [adenhq.com](https://adenhq.com) for complete documentation, examples, and guides.
 
-[![Hive Demo](https://img.youtube.com/vi/XDOG9fOaLjU/maxresdefault.jpg)](https://www.youtube.com/watch?v=XDOG9fOaLjU)
+https://github.com/user-attachments/assets/aad3a035-e7b3-4cac-b13d-4a83c7002c30
+
 
 ## Who Is Hive For?
 
-Hive is designed for developers and teams who want to build **production-grade AI agents** without manually wiring complex workflows.
+Hive is designed for developers and teams who want to build many **autonomous AI agents** fast without manually wiring complex workflows.
 
 Hive is a good fit if you:
 
@@ -84,7 +85,7 @@ Use Hive when you need:
 - An LLM provider that powers the agents
 - **ripgrep (optional, recommended on Windows):** The `search_files` tool uses ripgrep for faster file search. If not installed, a Python fallback is used. On Windows: `winget install BurntSushi.ripgrep` or `scoop install ripgrep`
 
-> **Note for Windows Users:** It is strongly recommended to use **WSL (Windows Subsystem for Linux)** or **Git Bash** to run this framework. Some core automation scripts may not execute correctly in standard Command Prompt or PowerShell.
+> **Windows Users:** Native Windows is supported via `quickstart.ps1` and `hive.ps1`. Run these in PowerShell 5.1+. WSL is also an option but not required.
 
 ### Installation
 
@@ -111,39 +112,36 @@ This sets up:
 - **LLM provider** - Interactive default model configuration
 - All required Python dependencies with `uv`
 
-- At last, it will initiate the open hive interface in your browser
+- Finally, it will open the Hive interface in your browser
 
 > **Tip:** To reopen the dashboard later, run `hive open` from the project directory.
 
-<img width="2500" height="1214" alt="home-screen" src="https://github.com/user-attachments/assets/134d897f-5e75-4874-b00b-e0505f6b45c4" />
-
 ### Build Your First Agent
 
-Type the agent you want to build in the home input box
+Type the agent you want to build in the home input box. The queen is going to ask you questions and work out a solution with you.
 
 <img width="2500" height="1214" alt="Image" src="https://github.com/user-attachments/assets/1ce19141-a78b-46f5-8d64-dbf987e048f4" />
 
 ### Use Template Agents
 
-Click "Try a sample agent" and check the templates. You can run a templates directly or choose to build your version on top of the existing template.
+Click "Try a sample agent" and check the templates. You can run a template directly or choose to build your version on top of the existing template.
 
 ### Run Agents
 
-Now you can run an agent by selectiing the agent (either an existing agent or example agent). You can click the Run button on the top left, or talk to the queen agent and it can run the agent for you.
+Now you can run an agent by selecting the agent (either an existing agent or example agent). You can click the Run button on the top left, or talk to the queen agent and it can run the agent for you.
 
-<img width="2500" height="1214" alt="Image" src="https://github.com/user-attachments/assets/71c38206-2ad5-49aa-bde8-6698d0bc55f5" />
+<img width="2549" height="1174" alt="Screenshot 2026-03-12 at 9 27 36 PM" src="https://github.com/user-attachments/assets/7c7d30fa-9ceb-4c23-95af-b1caa405547d" />
 
 ## Features
 
 - **Browser-Use** - Control the browser on your computer to achieve hard tasks
-- **Parallel Execution** - Execute the generated graph in parallel. This way you can have multiple agent compelteing the jobs for you
+- **Parallel Execution** - Execute the generated graph in parallel. This way you can have multiple agents completing the jobs for you
 - **[Goal-Driven Generation](docs/key_concepts/goals_outcome.md)** - Define objectives in natural language; the coding agent generates the agent graph and connection code to achieve them
 - **[Adaptiveness](docs/key_concepts/evolution.md)** - Framework captures failures, calibrates according to the objectives, and evolves the agent graph
 - **[Dynamic Node Connections](docs/key_concepts/graph.md)** - No predefined edges; connection code is generated by any capable LLM based on your goals
 - **SDK-Wrapped Nodes** - Every node gets shared memory, local RLM memory, monitoring, tools, and LLM access out of the box
 - **[Human-in-the-Loop](docs/key_concepts/graph.md#human-in-the-loop)** - Intervention nodes that pause execution for human input with configurable timeouts and escalation
 - **Real-time Observability** - WebSocket streaming for live monitoring of agent execution, decisions, and node-to-node communication
-- **Production-Ready** - Self-hostable, built for scale and reliability
 
 ## Integration
 
@@ -392,10 +390,6 @@ Hive generates your entire agent system from natural language goals using a codi
 
 Yes, Hive is fully open-source under the Apache License 2.0. We actively encourage community contributions and collaboration.
 
-**Q: Can Hive handle complex, production-scale use cases?**
-
-Yes. Hive is explicitly designed for production environments with features like automatic failure recovery, real-time observability, cost controls, and horizontal scaling support. The framework handles both simple automations and complex multi-agent workflows.
-
 **Q: Does Hive support human-in-the-loop workflows?**
 
 Yes, Hive fully supports [human-in-the-loop](docs/key_concepts/graph.md#human-in-the-loop) workflows through intervention nodes that pause execution for human input. These include configurable timeouts and escalation policies, allowing seamless collaboration between human experts and AI agents.
@@ -420,6 +414,16 @@ Visit [docs.adenhq.com](https://docs.adenhq.com/) for complete guides, API refer
 
 Contributions are welcome! Fork the repository, create your feature branch, implement your changes, and submit a pull request. See [CONTRIBUTING.md](CONTRIBUTING.md) for detailed guidelines.
 
+## Star History
+
+<a href="https://star-history.com/#aden-hive/hive&Date">
+ <picture>
+   <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/svg?repos=aden-hive/hive&type=Date&theme=dark" />
+   <source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/svg?repos=aden-hive/hive&type=Date" />
+   <img alt="Star History Chart" src="https://api.star-history.com/svg?repos=aden-hive/hive&type=Date" />
+ </picture>
+</a>
+
 ---
 
 <p align="center">

SECURITY.md

@@ -39,8 +39,8 @@ We consider security research conducted in accordance with this policy to be:
 ## Security Best Practices for Users
 
 1. **Keep Updated**: Always run the latest version
-2. **Secure Configuration**: Review `config.yaml` settings, especially in production
-3. **Environment Variables**: Never commit `.env` files or `config.yaml` with secrets
+2. **Secure Configuration**: Review your `~/.hive/configuration.json`, `.mcp.json`, and environment variable settings, especially in production
+3. **Environment Variables**: Never commit `.env` files or any configuration files that contain secrets
 4. **Network Security**: Use HTTPS in production, configure firewalls appropriately
 5. **Database Security**: Use strong passwords, limit network access
 

commit_message.txt

@@ -1,31 +0,0 @@
-perf: reduce subprocess spawning in quickstart scripts (#4427)
-
-## Problem
-Windows process creation (CreateProcess) is 10-100x slower than Linux fork/exec.
-The quickstart scripts were spawning 4+ separate `uv run python -c "import X"` 
-processes to verify imports, adding ~600ms overhead on Windows.
-
-## Solution
-Consolidated all import checks into a single batch script that checks multiple 
-modules in one subprocess call, reducing spawn overhead by ~75%.
-
-## Changes
-- **New**: `scripts/check_requirements.py` - Batched import checker
-- **New**: `scripts/test_check_requirements.py` - Test suite  
-- **New**: `scripts/benchmark_quickstart.ps1` - Performance benchmark tool
-- **Modified**: `quickstart.ps1` - Updated import verification (2 sections)
-- **Modified**: `quickstart.sh` - Updated import verification
-
-## Performance Impact
-**Benchmark results on Windows:**
-- Before: ~19.8 seconds for import checks
-- After: ~4.9 seconds for import checks
-- **Improvement: 14.9 seconds saved (75.2% faster)**
-
-## Testing
-- ✅ All functional tests pass (`scripts/test_check_requirements.py`)
-- ✅ Quickstart scripts work correctly on Windows
-- ✅ Error handling verified (invalid imports reported correctly)
-- ✅ Performance benchmark confirms 75%+ improvement
-
-Fixes #4427

core/demos/event_loop_wss_demo.py

@@ -1,740 +0,0 @@
-#!/usr/bin/env python3
-"""
-EventLoopNode WebSocket Demo
-
-Real LLM, real FileConversationStore, real EventBus.
-Streams EventLoopNode execution to a browser via WebSocket.
-
-Usage:
-    cd /home/timothy/oss/hive/core
-    python demos/event_loop_wss_demo.py
-
-    Then open http://localhost:8765 in your browser.
-"""
-
-import asyncio
-import json
-import logging
-import sys
-import tempfile
-from http import HTTPStatus
-from pathlib import Path
-
-import httpx
-import websockets
-from bs4 import BeautifulSoup
-from websockets.http11 import Request, Response
-
-# Add core, tools, and hive root to path
-_CORE_DIR = Path(__file__).resolve().parent.parent
-_HIVE_DIR = _CORE_DIR.parent
-sys.path.insert(0, str(_CORE_DIR))  # framework.*
-sys.path.insert(0, str(_HIVE_DIR / "tools" / "src"))  # aden_tools.*
-sys.path.insert(0, str(_HIVE_DIR))  # core.framework.* (for aden_tools imports)
-
-import os  # noqa: E402
-
-from aden_tools.credentials import CREDENTIAL_SPECS, CredentialStoreAdapter  # noqa: E402
-from core.framework.credentials import CredentialStore  # noqa: E402
-
-from framework.credentials.storage import (  # noqa: E402
-    CompositeStorage,
-    EncryptedFileStorage,
-    EnvVarStorage,
-)
-from framework.graph.event_loop_node import EventLoopNode, LoopConfig  # noqa: E402
-from framework.graph.node import NodeContext, NodeSpec, SharedMemory  # noqa: E402
-from framework.llm.litellm import LiteLLMProvider  # noqa: E402
-from framework.llm.provider import Tool  # noqa: E402
-from framework.runner.tool_registry import ToolRegistry  # noqa: E402
-from framework.runtime.core import Runtime  # noqa: E402
-from framework.runtime.event_bus import EventBus, EventType  # noqa: E402
-from framework.storage.conversation_store import FileConversationStore  # noqa: E402
-
-logging.basicConfig(level=logging.INFO, format="%(asctime)s %(name)s %(message)s")
-logger = logging.getLogger("demo")
-
-# -------------------------------------------------------------------------
-# Persistent state (shared across WebSocket connections)
-# -------------------------------------------------------------------------
-
-STORE_DIR = Path(tempfile.mkdtemp(prefix="hive_demo_"))
-STORE = FileConversationStore(STORE_DIR / "conversation")
-RUNTIME = Runtime(STORE_DIR / "runtime")
-LLM = LiteLLMProvider(model="claude-sonnet-4-5-20250929")
-
-# -------------------------------------------------------------------------
-# Tool Registry — real tools via ToolRegistry (same pattern as GraphExecutor)
-# -------------------------------------------------------------------------
-
-TOOL_REGISTRY = ToolRegistry()
-
-# Credential store: Aden sync (OAuth2 tokens) + encrypted files + env var fallback
-_env_mapping = {name: spec.env_var for name, spec in CREDENTIAL_SPECS.items()}
-_local_storage = CompositeStorage(
-    primary=EncryptedFileStorage(),
-    fallbacks=[EnvVarStorage(env_mapping=_env_mapping)],
-)
-
-if os.environ.get("ADEN_API_KEY"):
-    try:
-        from framework.credentials.aden import (  # noqa: E402
-            AdenCachedStorage,
-            AdenClientConfig,
-            AdenCredentialClient,
-            AdenSyncProvider,
-        )
-
-        _client = AdenCredentialClient(AdenClientConfig(base_url="https://api.adenhq.com"))
-        _provider = AdenSyncProvider(client=_client)
-        _storage = AdenCachedStorage(
-            local_storage=_local_storage,
-            aden_provider=_provider,
-        )
-        _cred_store = CredentialStore(storage=_storage, providers=[_provider], auto_refresh=True)
-        _synced = _provider.sync_all(_cred_store)
-        logger.info("Synced %d credentials from Aden", _synced)
-    except Exception as e:
-        logger.warning("Aden sync unavailable: %s", e)
-        _cred_store = CredentialStore(storage=_local_storage)
-else:
-    logger.info("ADEN_API_KEY not set, using local credential storage")
-    _cred_store = CredentialStore(storage=_local_storage)
-
-CREDENTIALS = CredentialStoreAdapter(_cred_store)
-
-# Debug: log which credentials resolved
-for _name in ["brave_search", "hubspot", "anthropic"]:
-    _val = CREDENTIALS.get(_name)
-    if _val:
-        logger.debug("credential %s: OK (len=%d)", _name, len(_val))
-    else:
-        logger.debug("credential %s: not found", _name)
-
-# --- web_search (Brave Search API) ---
-
-TOOL_REGISTRY.register(
-    name="web_search",
-    tool=Tool(
-        name="web_search",
-        description=(
-            "Search the web for current information. "
-            "Returns titles, URLs, and snippets from search results."
-        ),
-        parameters={
-            "type": "object",
-            "properties": {
-                "query": {
-                    "type": "string",
-                    "description": "The search query (1-500 characters)",
-                },
-                "num_results": {
-                    "type": "integer",
-                    "description": "Number of results to return (1-20, default 10)",
-                },
-            },
-            "required": ["query"],
-        },
-    ),
-    executor=lambda inputs: _exec_web_search(inputs),
-)
-
-
-def _exec_web_search(inputs: dict) -> dict:
-    api_key = CREDENTIALS.get("brave_search")
-    if not api_key:
-        return {"error": "brave_search credential not configured"}
-    query = inputs.get("query", "")
-    num_results = min(inputs.get("num_results", 10), 20)
-    resp = httpx.get(
-        "https://api.search.brave.com/res/v1/web/search",
-        params={"q": query, "count": num_results},
-        headers={"X-Subscription-Token": api_key, "Accept": "application/json"},
-        timeout=30.0,
-    )
-    if resp.status_code != 200:
-        return {"error": f"Brave API HTTP {resp.status_code}"}
-    data = resp.json()
-    results = [
-        {
-            "title": item.get("title", ""),
-            "url": item.get("url", ""),
-            "snippet": item.get("description", ""),
-        }
-        for item in data.get("web", {}).get("results", [])[:num_results]
-    ]
-    return {"query": query, "results": results, "total": len(results)}
-
-
-# --- web_scrape (httpx + BeautifulSoup, no playwright for sync compat) ---
-
-TOOL_REGISTRY.register(
-    name="web_scrape",
-    tool=Tool(
-        name="web_scrape",
-        description=(
-            "Scrape and extract text content from a webpage URL. "
-            "Returns the page title and main text content."
-        ),
-        parameters={
-            "type": "object",
-            "properties": {
-                "url": {
-                    "type": "string",
-                    "description": "URL of the webpage to scrape",
-                },
-                "max_length": {
-                    "type": "integer",
-                    "description": "Maximum text length (default 50000)",
-                },
-            },
-            "required": ["url"],
-        },
-    ),
-    executor=lambda inputs: _exec_web_scrape(inputs),
-)
-
-_SCRAPE_HEADERS = {
-    "User-Agent": (
-        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
-        "AppleWebKit/537.36 (KHTML, like Gecko) "
-        "Chrome/131.0.0.0 Safari/537.36"
-    ),
-    "Accept": "text/html,application/xhtml+xml",
-}
-
-
-def _exec_web_scrape(inputs: dict) -> dict:
-    url = inputs.get("url", "")
-    max_length = max(1000, min(inputs.get("max_length", 50000), 500000))
-    if not url.startswith(("http://", "https://")):
-        url = "https://" + url
-    try:
-        resp = httpx.get(url, timeout=30.0, follow_redirects=True, headers=_SCRAPE_HEADERS)
-        if resp.status_code != 200:
-            return {"error": f"HTTP {resp.status_code}"}
-        soup = BeautifulSoup(resp.text, "html.parser")
-        for tag in soup(["script", "style", "nav", "footer", "header", "aside", "noscript"]):
-            tag.decompose()
-        title = soup.title.get_text(strip=True) if soup.title else ""
-        main = (
-            soup.find("article")
-            or soup.find("main")
-            or soup.find(attrs={"role": "main"})
-            or soup.find("body")
-        )
-        text = main.get_text(separator=" ", strip=True) if main else ""
-        text = " ".join(text.split())
-        if len(text) > max_length:
-            text = text[:max_length] + "..."
-        return {"url": url, "title": title, "content": text, "length": len(text)}
-    except httpx.TimeoutException:
-        return {"error": "Request timed out"}
-    except Exception as e:
-        return {"error": f"Scrape failed: {e}"}
-
-
-# --- HubSpot CRM tools (optional, requires HUBSPOT_ACCESS_TOKEN) ---
-
-_HUBSPOT_API = "https://api.hubapi.com"
-
-
-def _hubspot_headers() -> dict | None:
-    token = CREDENTIALS.get("hubspot")
-    if token:
-        logger.debug("HubSpot token: %s...%s (len=%d)", token[:8], token[-4:], len(token))
-    else:
-        logger.debug("HubSpot token: not found")
-    if not token:
-        return None
-    return {
-        "Authorization": f"Bearer {token}",
-        "Content-Type": "application/json",
-        "Accept": "application/json",
-    }
-
-
-def _exec_hubspot_search(inputs: dict) -> dict:
-    headers = _hubspot_headers()
-    if not headers:
-        return {"error": "HUBSPOT_ACCESS_TOKEN not set"}
-    object_type = inputs.get("object_type", "contacts")
-    query = inputs.get("query", "")
-    limit = min(inputs.get("limit", 10), 100)
-    body: dict = {"limit": limit}
-    if query:
-        body["query"] = query
-    try:
-        resp = httpx.post(
-            f"{_HUBSPOT_API}/crm/v3/objects/{object_type}/search",
-            headers=headers,
-            json=body,
-            timeout=30.0,
-        )
-        if resp.status_code != 200:
-            return {"error": f"HubSpot API HTTP {resp.status_code}: {resp.text[:200]}"}
-        return resp.json()
-    except httpx.TimeoutException:
-        return {"error": "Request timed out"}
-    except Exception as e:
-        return {"error": f"HubSpot error: {e}"}
-
-
-TOOL_REGISTRY.register(
-    name="hubspot_search",
-    tool=Tool(
-        name="hubspot_search",
-        description=(
-            "Search HubSpot CRM objects (contacts, companies, or deals). "
-            "Returns matching records with their properties."
-        ),
-        parameters={
-            "type": "object",
-            "properties": {
-                "object_type": {
-                    "type": "string",
-                    "description": "CRM object type: 'contacts', 'companies', or 'deals'",
-                },
-                "query": {
-                    "type": "string",
-                    "description": "Search query (name, email, domain, etc.)",
-                },
-                "limit": {
-                    "type": "integer",
-                    "description": "Max results (1-100, default 10)",
-                },
-            },
-            "required": ["object_type"],
-        },
-    ),
-    executor=lambda inputs: _exec_hubspot_search(inputs),
-)
-
-logger.info(
-    "ToolRegistry loaded: %s",
-    ", ".join(TOOL_REGISTRY.get_registered_names()),
-)
-
-
-# -------------------------------------------------------------------------
-# HTML page (embedded)
-# -------------------------------------------------------------------------
-
-HTML_PAGE = (  # noqa: E501
-    """<!DOCTYPE html>
-<html lang="en">
-<head>
-<meta charset="utf-8">
-<meta name="viewport" content="width=device-width, initial-scale=1">
-<title>EventLoopNode Live Demo</title>
-<style>
-  * { box-sizing: border-box; margin: 0; padding: 0; }
-  body {
-    font-family: 'SF Mono', 'Fira Code', monospace;
-    background: #0d1117; color: #c9d1d9;
-    height: 100vh; display: flex; flex-direction: column;
-  }
-  header {
-    background: #161b22; padding: 12px 20px;
-    border-bottom: 1px solid #30363d;
-    display: flex; align-items: center; gap: 16px;
-  }
-  header h1 { font-size: 16px; color: #58a6ff; font-weight: 600; }
-  .status {
-    font-size: 12px; padding: 3px 10px; border-radius: 12px;
-    background: #21262d; color: #8b949e;
-  }
-  .status.running { background: #1a4b2e; color: #3fb950; }
-  .status.done { background: #1a3a5c; color: #58a6ff; }
-  .status.error { background: #4b1a1a; color: #f85149; }
-  .chat { flex: 1; overflow-y: auto; padding: 16px; }
-  .msg {
-    margin: 8px 0; padding: 10px 14px; border-radius: 8px;
-    line-height: 1.6; white-space: pre-wrap; word-wrap: break-word;
-  }
-  .msg.user { background: #1a3a5c; color: #58a6ff; }
-  .msg.assistant { background: #161b22; color: #c9d1d9; }
-  .msg.event {
-    background: transparent; color: #8b949e; font-size: 11px;
-    padding: 4px 14px; border-left: 3px solid #30363d;
-  }
-  .msg.event.loop { border-left-color: #58a6ff; }
-  .msg.event.tool { border-left-color: #d29922; }
-  .msg.event.stall { border-left-color: #f85149; }
-  .input-bar {
-    padding: 12px 16px; background: #161b22;
-    border-top: 1px solid #30363d; display: flex; gap: 8px;
-  }
-  .input-bar input {
-    flex: 1; background: #0d1117; border: 1px solid #30363d;
-    color: #c9d1d9; padding: 8px 12px; border-radius: 6px;
-    font-family: inherit; font-size: 14px; outline: none;
-  }
-  .input-bar input:focus { border-color: #58a6ff; }
-  .input-bar button {
-    background: #238636; color: #fff; border: none;
-    padding: 8px 20px; border-radius: 6px; cursor: pointer;
-    font-family: inherit; font-weight: 600;
-  }
-  .input-bar button:hover { background: #2ea043; }
-  .input-bar button:disabled {
-    background: #21262d; color: #484f58; cursor: not-allowed;
-  }
-  .input-bar button.clear { background: #da3633; }
-  .input-bar button.clear:hover { background: #f85149; }
-</style>
-</head>
-<body>
-  <header>
-    <h1>EventLoopNode Live</h1>
-    <span id="status" class="status">Idle</span>
-    <span id="iter" class="status" style="display:none">Step 0</span>
-  </header>
-  <div id="chat" class="chat"></div>
-  <div class="input-bar">
-    <input id="input" type="text"
-           placeholder="Ask anything..." autofocus />
-    <button id="go" onclick="run()">Send</button>
-    <button class="clear"
-            onclick="clearConversation()">Clear</button>
-  </div>
-
-<script>
-let ws = null;
-let currentAssistantEl = null;
-let iterCount = 0;
-const chat = document.getElementById('chat');
-const status = document.getElementById('status');
-const iterEl = document.getElementById('iter');
-const goBtn = document.getElementById('go');
-const inputEl = document.getElementById('input');
-
-inputEl.addEventListener('keydown', e => {
-  if (e.key === 'Enter') run();
-});
-
-function setStatus(text, cls) {
-  status.textContent = text;
-  status.className = 'status ' + cls;
-}
-
-function addMsg(text, cls) {
-  const el = document.createElement('div');
-  el.className = 'msg ' + cls;
-  el.textContent = text;
-  chat.appendChild(el);
-  chat.scrollTop = chat.scrollHeight;
-  return el;
-}
-
-function connect() {
-  ws = new WebSocket('ws://' + location.host + '/ws');
-  ws.onopen = () => {
-    setStatus('Ready', 'done');
-    goBtn.disabled = false;
-  };
-  ws.onmessage = handleEvent;
-  ws.onerror = () => { setStatus('Error', 'error'); };
-  ws.onclose = () => {
-    setStatus('Reconnecting...', '');
-    goBtn.disabled = true;
-    setTimeout(connect, 2000);
-  };
-}
-
-function handleEvent(msg) {
-  const evt = JSON.parse(msg.data);
-
-  if (evt.type === 'llm_text_delta') {
-    if (currentAssistantEl) {
-      currentAssistantEl.textContent += evt.content;
-      chat.scrollTop = chat.scrollHeight;
-    }
-  }
-  else if (evt.type === 'ready') {
-    setStatus('Ready', 'done');
-    if (currentAssistantEl && !currentAssistantEl.textContent)
-      currentAssistantEl.remove();
-    goBtn.disabled = false;
-  }
-  else if (evt.type === 'node_loop_iteration') {
-    iterCount = evt.iteration || (iterCount + 1);
-    iterEl.textContent = 'Step ' + iterCount;
-    iterEl.style.display = '';
-  }
-  else if (evt.type === 'tool_call_started') {
-    var info = evt.tool_name + '('
-      + JSON.stringify(evt.tool_input).slice(0, 120) + ')';
-    addMsg('TOOL  ' + info, 'event tool');
-  }
-  else if (evt.type === 'tool_call_completed') {
-    var preview = (evt.result || '').slice(0, 200);
-    var cls = evt.is_error ? 'stall' : 'tool';
-    addMsg('RESULT  ' + evt.tool_name + ': ' + preview,
-           'event ' + cls);
-    currentAssistantEl = addMsg('', 'assistant');
-  }
-  else if (evt.type === 'result') {
-    setStatus('Session ended', evt.success ? 'done' : 'error');
-    if (evt.error) addMsg('ERROR  ' + evt.error, 'event stall');
-    if (currentAssistantEl && !currentAssistantEl.textContent)
-      currentAssistantEl.remove();
-    goBtn.disabled = false;
-  }
-  else if (evt.type === 'node_stalled') {
-    addMsg('STALLED  ' + evt.reason, 'event stall');
-  }
-  else if (evt.type === 'cleared') {
-    chat.innerHTML = '';
-    iterCount = 0;
-    iterEl.textContent = 'Step 0';
-    iterEl.style.display = 'none';
-    setStatus('Ready', 'done');
-    goBtn.disabled = false;
-  }
-}
-
-function run() {
-  const text = inputEl.value.trim();
-  if (!text || !ws || ws.readyState !== 1) return;
-  addMsg(text, 'user');
-  currentAssistantEl = addMsg('', 'assistant');
-  inputEl.value = '';
-  setStatus('Running', 'running');
-  goBtn.disabled = true;
-  ws.send(JSON.stringify({ topic: text }));
-}
-
-function clearConversation() {
-  if (ws && ws.readyState === 1) {
-    ws.send(JSON.stringify({ command: 'clear' }));
-  }
-}
-
-connect();
-</script>
-</body>
-</html>"""
-)
-
-
-# -------------------------------------------------------------------------
-# WebSocket handler
-# -------------------------------------------------------------------------
-
-
-async def handle_ws(websocket):
-    """Persistent WebSocket: long-lived EventLoopNode with client_facing blocking."""
-    global STORE
-
-    # -- Event forwarding (WebSocket ← EventBus) ----------------------------
-    bus = EventBus()
-
-    async def forward_event(event):
-        try:
-            payload = {"type": event.type.value, **event.data}
-            if event.node_id:
-                payload["node_id"] = event.node_id
-            await websocket.send(json.dumps(payload))
-        except Exception:
-            pass
-
-    bus.subscribe(
-        event_types=[
-            EventType.NODE_LOOP_STARTED,
-            EventType.NODE_LOOP_ITERATION,
-            EventType.NODE_LOOP_COMPLETED,
-            EventType.LLM_TEXT_DELTA,
-            EventType.TOOL_CALL_STARTED,
-            EventType.TOOL_CALL_COMPLETED,
-            EventType.NODE_STALLED,
-        ],
-        handler=forward_event,
-    )
-
-    # -- Per-connection state -----------------------------------------------
-    node = None
-    loop_task = None
-
-    tools = list(TOOL_REGISTRY.get_tools().values())
-    tool_executor = TOOL_REGISTRY.get_executor()
-
-    node_spec = NodeSpec(
-        id="assistant",
-        name="Chat Assistant",
-        description="A conversational assistant that remembers context across messages",
-        node_type="event_loop",
-        client_facing=True,
-        system_prompt=(
-            "You are a helpful assistant with access to tools. "
-            "You can search the web, scrape webpages, and query HubSpot CRM. "
-            "Use tools when the user asks for current information or external data. "
-            "You have full conversation history, so you can reference previous messages."
-        ),
-    )
-
-    # -- Ready callback: subscribe to CLIENT_INPUT_REQUESTED on the bus ---
-    async def on_input_requested(event):
-        try:
-            await websocket.send(json.dumps({"type": "ready"}))
-        except Exception:
-            pass
-
-    bus.subscribe(
-        event_types=[EventType.CLIENT_INPUT_REQUESTED],
-        handler=on_input_requested,
-    )
-
-    async def start_loop(first_message: str):
-        """Create an EventLoopNode and run it as a background task."""
-        nonlocal node, loop_task
-
-        memory = SharedMemory()
-        ctx = NodeContext(
-            runtime=RUNTIME,
-            node_id="assistant",
-            node_spec=node_spec,
-            memory=memory,
-            input_data={},
-            llm=LLM,
-            available_tools=tools,
-        )
-        node = EventLoopNode(
-            event_bus=bus,
-            config=LoopConfig(max_iterations=10_000, max_history_tokens=32_000),
-            conversation_store=STORE,
-            tool_executor=tool_executor,
-        )
-        await node.inject_event(first_message)
-
-        async def _run():
-            try:
-                result = await node.execute(ctx)
-                try:
-                    await websocket.send(
-                        json.dumps(
-                            {
-                                "type": "result",
-                                "success": result.success,
-                                "output": result.output,
-                                "error": result.error,
-                                "tokens": result.tokens_used,
-                            }
-                        )
-                    )
-                except Exception:
-                    pass
-                logger.info(f"Loop ended: success={result.success}, tokens={result.tokens_used}")
-            except websockets.exceptions.ConnectionClosed:
-                logger.info("Loop stopped: WebSocket closed")
-            except Exception as e:
-                logger.exception("Loop error")
-                try:
-                    await websocket.send(
-                        json.dumps(
-                            {
-                                "type": "result",
-                                "success": False,
-                                "error": str(e),
-                                "output": {},
-                            }
-                        )
-                    )
-                except Exception:
-                    pass
-
-        loop_task = asyncio.create_task(_run())
-
-    async def stop_loop():
-        """Signal the node and wait for the loop task to finish."""
-        nonlocal node, loop_task
-        if loop_task and not loop_task.done():
-            if node:
-                node.signal_shutdown()
-            try:
-                await asyncio.wait_for(loop_task, timeout=5.0)
-            except (TimeoutError, asyncio.CancelledError):
-                loop_task.cancel()
-        node = None
-        loop_task = None
-
-    # -- Message loop (runs for the lifetime of this WebSocket) -------------
-    try:
-        async for raw in websocket:
-            try:
-                msg = json.loads(raw)
-            except Exception:
-                continue
-
-            # Clear command
-            if msg.get("command") == "clear":
-                import shutil
-
-                await stop_loop()
-                await STORE.close()
-                conv_dir = STORE_DIR / "conversation"
-                if conv_dir.exists():
-                    shutil.rmtree(conv_dir)
-                STORE = FileConversationStore(conv_dir)
-                await websocket.send(json.dumps({"type": "cleared"}))
-                logger.info("Conversation cleared")
-                continue
-
-            topic = msg.get("topic", "")
-            if not topic:
-                continue
-
-            if node is None:
-                # First message — spin up the loop
-                logger.info(f"Starting persistent loop: {topic}")
-                await start_loop(topic)
-            else:
-                # Subsequent message — inject into the running loop
-                logger.info(f"Injecting message: {topic}")
-                await node.inject_event(topic)
-
-    except websockets.exceptions.ConnectionClosed:
-        pass
-    finally:
-        await stop_loop()
-        logger.info("WebSocket closed, loop stopped")
-
-
-# -------------------------------------------------------------------------
-# HTTP handler for serving the HTML page
-# -------------------------------------------------------------------------
-
-
-async def process_request(connection, request: Request):
-    """Serve HTML on GET /, upgrade to WebSocket on /ws."""
-    if request.path == "/ws":
-        return None  # let websockets handle the upgrade
-    # Serve the HTML page for any other path
-    return Response(
-        HTTPStatus.OK,
-        "OK",
-        websockets.Headers({"Content-Type": "text/html; charset=utf-8"}),
-        HTML_PAGE.encode(),
-    )
-
-
-# -------------------------------------------------------------------------
-# Main
-# -------------------------------------------------------------------------
-
-
-async def main():
-    port = 8765
-    async with websockets.serve(
-        handle_ws,
-        "0.0.0.0",
-        port,
-        process_request=process_request,
-    ):
-        logger.info(f"Demo running at http://localhost:{port}")
-        logger.info("Open in your browser and enter a topic to research.")
-        await asyncio.Future()  # run forever
-
-
-if __name__ == "__main__":
-    asyncio.run(main())

core/demos/handoff_demo.py

@@ -1,930 +0,0 @@
-#!/usr/bin/env python3
-"""
-Two-Node ContextHandoff Demo
-
-Demonstrates ContextHandoff between two EventLoopNode instances:
-  Node A (Researcher) → ContextHandoff → Node B (Analyst)
-
-Real LLM, real FileConversationStore, real EventBus.
-Streams both nodes to a browser via WebSocket.
-
-Usage:
-    cd /home/timothy/oss/hive/core
-    python demos/handoff_demo.py
-
-    Then open http://localhost:8766 in your browser.
-"""
-
-import asyncio
-import json
-import logging
-import sys
-import tempfile
-from http import HTTPStatus
-from pathlib import Path
-
-import httpx
-import websockets
-from bs4 import BeautifulSoup
-from websockets.http11 import Request, Response
-
-# Add core, tools, and hive root to path
-_CORE_DIR = Path(__file__).resolve().parent.parent
-_HIVE_DIR = _CORE_DIR.parent
-sys.path.insert(0, str(_CORE_DIR))  # framework.*
-sys.path.insert(0, str(_HIVE_DIR / "tools" / "src"))  # aden_tools.*
-sys.path.insert(0, str(_HIVE_DIR))  # core.framework.* (for aden_tools imports)
-
-from aden_tools.credentials import CREDENTIAL_SPECS, CredentialStoreAdapter  # noqa: E402
-from core.framework.credentials import CredentialStore  # noqa: E402
-
-from framework.credentials.storage import (  # noqa: E402
-    CompositeStorage,
-    EncryptedFileStorage,
-    EnvVarStorage,
-)
-from framework.graph.context_handoff import ContextHandoff  # noqa: E402
-from framework.graph.conversation import NodeConversation  # noqa: E402
-from framework.graph.event_loop_node import EventLoopNode, LoopConfig  # noqa: E402
-from framework.graph.node import NodeContext, NodeSpec, SharedMemory  # noqa: E402
-from framework.llm.litellm import LiteLLMProvider  # noqa: E402
-from framework.llm.provider import Tool  # noqa: E402
-from framework.runner.tool_registry import ToolRegistry  # noqa: E402
-from framework.runtime.core import Runtime  # noqa: E402
-from framework.runtime.event_bus import EventBus, EventType  # noqa: E402
-from framework.storage.conversation_store import FileConversationStore  # noqa: E402
-
-logging.basicConfig(level=logging.INFO, format="%(asctime)s %(name)s %(message)s")
-logger = logging.getLogger("handoff_demo")
-
-# -------------------------------------------------------------------------
-# Persistent state
-# -------------------------------------------------------------------------
-
-STORE_DIR = Path(tempfile.mkdtemp(prefix="hive_handoff_"))
-RUNTIME = Runtime(STORE_DIR / "runtime")
-LLM = LiteLLMProvider(model="claude-sonnet-4-5-20250929")
-
-# -------------------------------------------------------------------------
-# Credentials
-# -------------------------------------------------------------------------
-
-# Composite credential store: encrypted files (primary) + env vars (fallback)
-_env_mapping = {name: spec.env_var for name, spec in CREDENTIAL_SPECS.items()}
-_composite = CompositeStorage(
-    primary=EncryptedFileStorage(),
-    fallbacks=[EnvVarStorage(env_mapping=_env_mapping)],
-)
-CREDENTIALS = CredentialStoreAdapter(CredentialStore(storage=_composite))
-
-for _name in ["brave_search", "hubspot"]:
-    _val = CREDENTIALS.get(_name)
-    if _val:
-        logger.debug("credential %s: OK (len=%d)", _name, len(_val))
-    else:
-        logger.debug("credential %s: not found", _name)
-
-# -------------------------------------------------------------------------
-# Tool Registry — web_search + web_scrape for Node A (Researcher)
-# -------------------------------------------------------------------------
-
-TOOL_REGISTRY = ToolRegistry()
-
-
-def _exec_web_search(inputs: dict) -> dict:
-    api_key = CREDENTIALS.get("brave_search")
-    if not api_key:
-        return {"error": "brave_search credential not configured"}
-    query = inputs.get("query", "")
-    num_results = min(inputs.get("num_results", 10), 20)
-    resp = httpx.get(
-        "https://api.search.brave.com/res/v1/web/search",
-        params={"q": query, "count": num_results},
-        headers={
-            "X-Subscription-Token": api_key,
-            "Accept": "application/json",
-        },
-        timeout=30.0,
-    )
-    if resp.status_code != 200:
-        return {"error": f"Brave API HTTP {resp.status_code}"}
-    data = resp.json()
-    results = [
-        {
-            "title": item.get("title", ""),
-            "url": item.get("url", ""),
-            "snippet": item.get("description", ""),
-        }
-        for item in data.get("web", {}).get("results", [])[:num_results]
-    ]
-    return {"query": query, "results": results, "total": len(results)}
-
-
-TOOL_REGISTRY.register(
-    name="web_search",
-    tool=Tool(
-        name="web_search",
-        description=(
-            "Search the web for current information. "
-            "Returns titles, URLs, and snippets from search results."
-        ),
-        parameters={
-            "type": "object",
-            "properties": {
-                "query": {
-                    "type": "string",
-                    "description": "The search query (1-500 characters)",
-                },
-                "num_results": {
-                    "type": "integer",
-                    "description": "Number of results (1-20, default 10)",
-                },
-            },
-            "required": ["query"],
-        },
-    ),
-    executor=lambda inputs: _exec_web_search(inputs),
-)
-
-_SCRAPE_HEADERS = {
-    "User-Agent": (
-        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
-        "AppleWebKit/537.36 (KHTML, like Gecko) "
-        "Chrome/131.0.0.0 Safari/537.36"
-    ),
-    "Accept": "text/html,application/xhtml+xml",
-}
-
-
-def _exec_web_scrape(inputs: dict) -> dict:
-    url = inputs.get("url", "")
-    max_length = max(1000, min(inputs.get("max_length", 50000), 500000))
-    if not url.startswith(("http://", "https://")):
-        url = "https://" + url
-    try:
-        resp = httpx.get(
-            url,
-            timeout=30.0,
-            follow_redirects=True,
-            headers=_SCRAPE_HEADERS,
-        )
-        if resp.status_code != 200:
-            return {"error": f"HTTP {resp.status_code}"}
-        soup = BeautifulSoup(resp.text, "html.parser")
-        for tag in soup(["script", "style", "nav", "footer", "header", "aside", "noscript"]):
-            tag.decompose()
-        title = soup.title.get_text(strip=True) if soup.title else ""
-        main = (
-            soup.find("article")
-            or soup.find("main")
-            or soup.find(attrs={"role": "main"})
-            or soup.find("body")
-        )
-        text = main.get_text(separator=" ", strip=True) if main else ""
-        text = " ".join(text.split())
-        if len(text) > max_length:
-            text = text[:max_length] + "..."
-        return {
-            "url": url,
-            "title": title,
-            "content": text,
-            "length": len(text),
-        }
-    except httpx.TimeoutException:
-        return {"error": "Request timed out"}
-    except Exception as e:
-        return {"error": f"Scrape failed: {e}"}
-
-
-TOOL_REGISTRY.register(
-    name="web_scrape",
-    tool=Tool(
-        name="web_scrape",
-        description=(
-            "Scrape and extract text content from a webpage URL. "
-            "Returns the page title and main text content."
-        ),
-        parameters={
-            "type": "object",
-            "properties": {
-                "url": {
-                    "type": "string",
-                    "description": "URL of the webpage to scrape",
-                },
-                "max_length": {
-                    "type": "integer",
-                    "description": "Maximum text length (default 50000)",
-                },
-            },
-            "required": ["url"],
-        },
-    ),
-    executor=lambda inputs: _exec_web_scrape(inputs),
-)
-
-logger.info(
-    "ToolRegistry loaded: %s",
-    ", ".join(TOOL_REGISTRY.get_registered_names()),
-)
-
-# -------------------------------------------------------------------------
-# Node Specs
-# -------------------------------------------------------------------------
-
-RESEARCHER_SPEC = NodeSpec(
-    id="researcher",
-    name="Researcher",
-    description="Researches a topic using web search and scraping tools",
-    node_type="event_loop",
-    input_keys=["topic"],
-    output_keys=["research_summary"],
-    system_prompt=(
-        "You are a thorough research assistant. Your job is to research "
-        "the given topic using the web_search and web_scrape tools.\n\n"
-        "1. Search for relevant information on the topic\n"
-        "2. Scrape 1-2 of the most promising URLs for details\n"
-        "3. Synthesize your findings into a comprehensive summary\n"
-        "4. Use set_output with key='research_summary' to save your "
-        "findings\n\n"
-        "Be thorough but efficient. Aim for 2-4 search/scrape calls, "
-        "then summarize and set_output."
-    ),
-)
-
-ANALYST_SPEC = NodeSpec(
-    id="analyst",
-    name="Analyst",
-    description="Analyzes research findings and provides insights",
-    node_type="event_loop",
-    input_keys=["context"],
-    output_keys=["analysis"],
-    system_prompt=(
-        "You are a strategic analyst. You receive research findings from "
-        "a previous researcher and must:\n\n"
-        "1. Identify key themes and patterns\n"
-        "2. Assess the reliability and significance of the findings\n"
-        "3. Provide actionable insights and recommendations\n"
-        "4. Use set_output with key='analysis' to save your analysis\n\n"
-        "Be concise but insightful. Focus on what matters most."
-    ),
-)
-
-
-# -------------------------------------------------------------------------
-# HTML page
-# -------------------------------------------------------------------------
-
-HTML_PAGE = (  # noqa: E501
-    """<!DOCTYPE html>
-<html lang="en">
-<head>
-<meta charset="utf-8">
-<meta name="viewport" content="width=device-width, initial-scale=1">
-<title>ContextHandoff Demo</title>
-<style>
-  * {
-    box-sizing: border-box;
-    margin: 0;
-    padding: 0;
-  }
-  body {
-    font-family: 'SF Mono', 'Fira Code', monospace;
-    background: #0d1117;
-    color: #c9d1d9;
-    height: 100vh;
-    display: flex;
-    flex-direction: column;
-  }
-  header {
-    background: #161b22;
-    padding: 12px 20px;
-    border-bottom: 1px solid #30363d;
-    display: flex;
-    align-items: center;
-    gap: 16px;
-  }
-  header h1 {
-    font-size: 16px;
-    color: #58a6ff;
-    font-weight: 600;
-  }
-  .badge {
-    font-size: 12px;
-    padding: 3px 10px;
-    border-radius: 12px;
-    background: #21262d;
-    color: #8b949e;
-  }
-  .badge.researcher {
-    background: #1a3a5c;
-    color: #58a6ff;
-  }
-  .badge.analyst {
-    background: #1a4b2e;
-    color: #3fb950;
-  }
-  .badge.handoff {
-    background: #3d1f00;
-    color: #d29922;
-  }
-  .badge.done {
-    background: #21262d;
-    color: #8b949e;
-  }
-  .badge.error {
-    background: #4b1a1a;
-    color: #f85149;
-  }
-  .chat {
-    flex: 1;
-    overflow-y: auto;
-    padding: 16px;
-  }
-  .msg {
-    margin: 8px 0;
-    padding: 10px 14px;
-    border-radius: 8px;
-    line-height: 1.6;
-    white-space: pre-wrap;
-    word-wrap: break-word;
-  }
-  .msg.user {
-    background: #1a3a5c;
-    color: #58a6ff;
-  }
-  .msg.assistant {
-    background: #161b22;
-    color: #c9d1d9;
-  }
-  .msg.assistant.analyst-msg {
-    border-left: 3px solid #3fb950;
-  }
-  .msg.event {
-    background: transparent;
-    color: #8b949e;
-    font-size: 11px;
-    padding: 4px 14px;
-    border-left: 3px solid #30363d;
-  }
-  .msg.event.loop {
-    border-left-color: #58a6ff;
-  }
-  .msg.event.tool {
-    border-left-color: #d29922;
-  }
-  .msg.event.stall {
-    border-left-color: #f85149;
-  }
-  .handoff-banner {
-    margin: 16px 0;
-    padding: 16px;
-    background: #1c1200;
-    border: 1px solid #d29922;
-    border-radius: 8px;
-    text-align: center;
-  }
-  .handoff-banner h3 {
-    color: #d29922;
-    font-size: 14px;
-    margin-bottom: 8px;
-  }
-  .handoff-banner p, .result-banner p {
-    color: #8b949e;
-    font-size: 12px;
-    line-height: 1.5;
-    max-height: 200px;
-    overflow-y: auto;
-    white-space: pre-wrap;
-    text-align: left;
-  }
-  .result-banner {
-    margin: 16px 0;
-    padding: 16px;
-    background: #0a2614;
-    border: 1px solid #3fb950;
-    border-radius: 8px;
-  }
-  .result-banner h3 {
-    color: #3fb950;
-    font-size: 14px;
-    margin-bottom: 8px;
-    text-align: center;
-  }
-  .result-banner .label {
-    color: #58a6ff;
-    font-size: 11px;
-    font-weight: 600;
-    margin-top: 10px;
-    margin-bottom: 2px;
-  }
-  .result-banner .tokens {
-    color: #484f58;
-    font-size: 11px;
-    text-align: center;
-    margin-top: 10px;
-  }
-  .input-bar {
-    padding: 12px 16px;
-    background: #161b22;
-    border-top: 1px solid #30363d;
-    display: flex;
-    gap: 8px;
-  }
-  .input-bar input {
-    flex: 1;
-    background: #0d1117;
-    border: 1px solid #30363d;
-    color: #c9d1d9;
-    padding: 8px 12px;
-    border-radius: 6px;
-    font-family: inherit;
-    font-size: 14px;
-    outline: none;
-  }
-  .input-bar input:focus {
-    border-color: #58a6ff;
-  }
-  .input-bar button {
-    background: #238636;
-    color: #fff;
-    border: none;
-    padding: 8px 20px;
-    border-radius: 6px;
-    cursor: pointer;
-    font-family: inherit;
-    font-weight: 600;
-  }
-  .input-bar button:hover {
-    background: #2ea043;
-  }
-  .input-bar button:disabled {
-    background: #21262d;
-    color: #484f58;
-    cursor: not-allowed;
-  }
-</style>
-</head>
-<body>
-  <header>
-    <h1>ContextHandoff Demo</h1>
-    <span id="phase" class="badge">Idle</span>
-    <span id="iter" class="badge" style="display:none">Step 0</span>
-  </header>
-  <div id="chat" class="chat"></div>
-  <div class="input-bar">
-    <input id="input" type="text"
-           placeholder="Enter a research topic..." autofocus />
-    <button id="go" onclick="run()">Research</button>
-  </div>
-
-<script>
-let ws = null;
-let currentAssistantEl = null;
-let iterCount = 0;
-let currentPhase = 'idle';
-const chat = document.getElementById('chat');
-const phase = document.getElementById('phase');
-const iterEl = document.getElementById('iter');
-const goBtn = document.getElementById('go');
-const inputEl = document.getElementById('input');
-
-inputEl.addEventListener('keydown', e => {
-  if (e.key === 'Enter') run();
-});
-
-function setPhase(text, cls) {
-  phase.textContent = text;
-  phase.className = 'badge ' + cls;
-  currentPhase = cls;
-}
-
-function addMsg(text, cls) {
-  const el = document.createElement('div');
-  el.className = 'msg ' + cls;
-  el.textContent = text;
-  chat.appendChild(el);
-  chat.scrollTop = chat.scrollHeight;
-  return el;
-}
-
-function addHandoffBanner(summary) {
-  const banner = document.createElement('div');
-  banner.className = 'handoff-banner';
-  const h3 = document.createElement('h3');
-  h3.textContent = 'Context Handoff: Researcher -> Analyst';
-  const p = document.createElement('p');
-  p.textContent = summary || 'Passing research context...';
-  banner.appendChild(h3);
-  banner.appendChild(p);
-  chat.appendChild(banner);
-  chat.scrollTop = chat.scrollHeight;
-}
-
-function addResultBanner(researcher, analyst, tokens) {
-  const banner = document.createElement('div');
-  banner.className = 'result-banner';
-  const h3 = document.createElement('h3');
-  h3.textContent = 'Pipeline Complete';
-  banner.appendChild(h3);
-
-  if (researcher && researcher.research_summary) {
-    const lbl = document.createElement('div');
-    lbl.className = 'label';
-    lbl.textContent = 'RESEARCH SUMMARY';
-    banner.appendChild(lbl);
-    const p = document.createElement('p');
-    p.textContent = researcher.research_summary;
-    banner.appendChild(p);
-  }
-
-  if (analyst && analyst.analysis) {
-    const lbl = document.createElement('div');
-    lbl.className = 'label';
-    lbl.textContent = 'ANALYSIS';
-    lbl.style.color = '#3fb950';
-    banner.appendChild(lbl);
-    const p = document.createElement('p');
-    p.textContent = analyst.analysis;
-    banner.appendChild(p);
-  }
-
-  if (tokens) {
-    const t = document.createElement('div');
-    t.className = 'tokens';
-    t.textContent = 'Total tokens: ' + tokens.toLocaleString();
-    banner.appendChild(t);
-  }
-
-  chat.appendChild(banner);
-  chat.scrollTop = chat.scrollHeight;
-}
-
-function connect() {
-  ws = new WebSocket('ws://' + location.host + '/ws');
-  ws.onopen = () => {
-    setPhase('Ready', 'done');
-    goBtn.disabled = false;
-  };
-  ws.onmessage = handleEvent;
-  ws.onerror = () => { setPhase('Error', 'error'); };
-  ws.onclose = () => {
-    setPhase('Reconnecting...', '');
-    goBtn.disabled = true;
-    setTimeout(connect, 2000);
-  };
-}
-
-function handleEvent(msg) {
-  const evt = JSON.parse(msg.data);
-
-  if (evt.type === 'phase') {
-    if (evt.phase === 'researcher') {
-      setPhase('Researcher', 'researcher');
-    } else if (evt.phase === 'handoff') {
-      setPhase('Handoff', 'handoff');
-    } else if (evt.phase === 'analyst') {
-      setPhase('Analyst', 'analyst');
-    }
-    iterCount = 0;
-    iterEl.style.display = 'none';
-  }
-  else if (evt.type === 'llm_text_delta') {
-    if (currentAssistantEl) {
-      currentAssistantEl.textContent += evt.content;
-      chat.scrollTop = chat.scrollHeight;
-    }
-  }
-  else if (evt.type === 'node_loop_iteration') {
-    iterCount = evt.iteration || (iterCount + 1);
-    iterEl.textContent = 'Step ' + iterCount;
-    iterEl.style.display = '';
-  }
-  else if (evt.type === 'tool_call_started') {
-    var info = evt.tool_name + '('
-      + JSON.stringify(evt.tool_input).slice(0, 120) + ')';
-    addMsg('TOOL  ' + info, 'event tool');
-  }
-  else if (evt.type === 'tool_call_completed') {
-    var preview = (evt.result || '').slice(0, 200);
-    var cls = evt.is_error ? 'stall' : 'tool';
-    addMsg(
-      'RESULT  ' + evt.tool_name + ': ' + preview,
-      'event ' + cls
-    );
-    var assistCls = currentPhase === 'analyst'
-      ? 'assistant analyst-msg' : 'assistant';
-    currentAssistantEl = addMsg('', assistCls);
-  }
-  else if (evt.type === 'handoff_context') {
-    addHandoffBanner(evt.summary);
-    var assistCls = 'assistant analyst-msg';
-    currentAssistantEl = addMsg('', assistCls);
-  }
-  else if (evt.type === 'node_result') {
-    if (evt.node_id === 'researcher') {
-      if (currentAssistantEl
-          && !currentAssistantEl.textContent) {
-        currentAssistantEl.remove();
-      }
-    }
-  }
-  else if (evt.type === 'done') {
-    setPhase('Done', 'done');
-    iterEl.style.display = 'none';
-    if (currentAssistantEl
-        && !currentAssistantEl.textContent) {
-      currentAssistantEl.remove();
-    }
-    currentAssistantEl = null;
-    addResultBanner(
-      evt.researcher, evt.analyst, evt.total_tokens
-    );
-    goBtn.disabled = false;
-    inputEl.placeholder = 'Enter another topic...';
-  }
-  else if (evt.type === 'error') {
-    setPhase('Error', 'error');
-    addMsg('ERROR  ' + evt.message, 'event stall');
-    goBtn.disabled = false;
-  }
-  else if (evt.type === 'node_stalled') {
-    addMsg('STALLED  ' + evt.reason, 'event stall');
-  }
-}
-
-function run() {
-  const text = inputEl.value.trim();
-  if (!text || !ws || ws.readyState !== 1) return;
-  chat.innerHTML = '';
-  addMsg(text, 'user');
-  currentAssistantEl = addMsg('', 'assistant');
-  inputEl.value = '';
-  goBtn.disabled = true;
-  ws.send(JSON.stringify({ topic: text }));
-}
-
-connect();
-</script>
-</body>
-</html>"""
-)
-
-
-# -------------------------------------------------------------------------
-# WebSocket handler — sequential Node A → Handoff → Node B
-# -------------------------------------------------------------------------
-
-
-async def handle_ws(websocket):
-    """Run the two-node handoff pipeline per user message."""
-    try:
-        async for raw in websocket:
-            try:
-                msg = json.loads(raw)
-            except Exception:
-                continue
-
-            topic = msg.get("topic", "")
-            if not topic:
-                continue
-
-            logger.info(f"Starting handoff pipeline for: {topic}")
-
-            try:
-                await _run_pipeline(websocket, topic)
-            except websockets.exceptions.ConnectionClosed:
-                logger.info("WebSocket closed during pipeline")
-                return
-            except Exception as e:
-                logger.exception("Pipeline error")
-                try:
-                    await websocket.send(json.dumps({"type": "error", "message": str(e)}))
-                except Exception:
-                    pass
-
-    except websockets.exceptions.ConnectionClosed:
-        pass
-
-
-async def _run_pipeline(websocket, topic: str):
-    """Execute: Node A (research) → ContextHandoff → Node B (analysis)."""
-    import shutil
-
-    # Fresh stores for each run
-    run_dir = Path(tempfile.mkdtemp(prefix="hive_run_", dir=STORE_DIR))
-    store_a = FileConversationStore(run_dir / "node_a")
-    store_b = FileConversationStore(run_dir / "node_b")
-
-    # Shared event bus
-    bus = EventBus()
-
-    async def forward_event(event):
-        try:
-            payload = {"type": event.type.value, **event.data}
-            if event.node_id:
-                payload["node_id"] = event.node_id
-            await websocket.send(json.dumps(payload))
-        except Exception:
-            pass
-
-    bus.subscribe(
-        event_types=[
-            EventType.NODE_LOOP_STARTED,
-            EventType.NODE_LOOP_ITERATION,
-            EventType.NODE_LOOP_COMPLETED,
-            EventType.LLM_TEXT_DELTA,
-            EventType.TOOL_CALL_STARTED,
-            EventType.TOOL_CALL_COMPLETED,
-            EventType.NODE_STALLED,
-        ],
-        handler=forward_event,
-    )
-
-    tools = list(TOOL_REGISTRY.get_tools().values())
-    tool_executor = TOOL_REGISTRY.get_executor()
-
-    # ---- Phase 1: Researcher ------------------------------------------------
-    await websocket.send(json.dumps({"type": "phase", "phase": "researcher"}))
-
-    node_a = EventLoopNode(
-        event_bus=bus,
-        judge=None,  # implicit judge: accept when output_keys filled
-        config=LoopConfig(
-            max_iterations=20,
-            max_tool_calls_per_turn=30,
-            max_history_tokens=32_000,
-        ),
-        conversation_store=store_a,
-        tool_executor=tool_executor,
-    )
-
-    ctx_a = NodeContext(
-        runtime=RUNTIME,
-        node_id="researcher",
-        node_spec=RESEARCHER_SPEC,
-        memory=SharedMemory(),
-        input_data={"topic": topic},
-        llm=LLM,
-        available_tools=tools,
-    )
-
-    result_a = await node_a.execute(ctx_a)
-    logger.info(
-        "Researcher done: success=%s, tokens=%s",
-        result_a.success,
-        result_a.tokens_used,
-    )
-
-    await websocket.send(
-        json.dumps(
-            {
-                "type": "node_result",
-                "node_id": "researcher",
-                "success": result_a.success,
-                "output": result_a.output,
-            }
-        )
-    )
-
-    if not result_a.success:
-        await websocket.send(
-            json.dumps(
-                {
-                    "type": "error",
-                    "message": f"Researcher failed: {result_a.error}",
-                }
-            )
-        )
-        return
-
-    # ---- Phase 2: Context Handoff -------------------------------------------
-    await websocket.send(json.dumps({"type": "phase", "phase": "handoff"}))
-
-    # Restore the researcher's conversation from store
-    conversation_a = await NodeConversation.restore(store_a)
-    if conversation_a is None:
-        await websocket.send(
-            json.dumps(
-                {
-                    "type": "error",
-                    "message": "Failed to restore researcher conversation",
-                }
-            )
-        )
-        return
-
-    handoff_engine = ContextHandoff(llm=LLM)
-    handoff_context = handoff_engine.summarize_conversation(
-        conversation=conversation_a,
-        node_id="researcher",
-        output_keys=["research_summary"],
-    )
-
-    formatted_handoff = ContextHandoff.format_as_input(handoff_context)
-    logger.info(
-        "Handoff: %d turns, ~%d tokens, keys=%s",
-        handoff_context.turn_count,
-        handoff_context.total_tokens_used,
-        list(handoff_context.key_outputs.keys()),
-    )
-
-    # Send handoff context to browser
-    await websocket.send(
-        json.dumps(
-            {
-                "type": "handoff_context",
-                "summary": handoff_context.summary[:500],
-                "turn_count": handoff_context.turn_count,
-                "tokens": handoff_context.total_tokens_used,
-                "key_outputs": handoff_context.key_outputs,
-            }
-        )
-    )
-
-    # ---- Phase 3: Analyst ---------------------------------------------------
-    await websocket.send(json.dumps({"type": "phase", "phase": "analyst"}))
-
-    node_b = EventLoopNode(
-        event_bus=bus,
-        judge=None,  # implicit judge
-        config=LoopConfig(
-            max_iterations=10,
-            max_tool_calls_per_turn=30,
-            max_history_tokens=32_000,
-        ),
-        conversation_store=store_b,
-    )
-
-    ctx_b = NodeContext(
-        runtime=RUNTIME,
-        node_id="analyst",
-        node_spec=ANALYST_SPEC,
-        memory=SharedMemory(),
-        input_data={"context": formatted_handoff},
-        llm=LLM,
-        available_tools=[],
-    )
-
-    result_b = await node_b.execute(ctx_b)
-    logger.info(
-        "Analyst done: success=%s, tokens=%s",
-        result_b.success,
-        result_b.tokens_used,
-    )
-
-    # ---- Done ---------------------------------------------------------------
-    await websocket.send(
-        json.dumps(
-            {
-                "type": "done",
-                "researcher": result_a.output,
-                "analyst": result_b.output,
-                "total_tokens": ((result_a.tokens_used or 0) + (result_b.tokens_used or 0)),
-            }
-        )
-    )
-
-    # Clean up temp stores
-    try:
-        shutil.rmtree(run_dir)
-    except Exception:
-        pass
-
-
-# -------------------------------------------------------------------------
-# HTTP handler
-# -------------------------------------------------------------------------
-
-
-async def process_request(connection, request: Request):
-    """Serve HTML on GET /, upgrade to WebSocket on /ws."""
-    if request.path == "/ws":
-        return None
-    return Response(
-        HTTPStatus.OK,
-        "OK",
-        websockets.Headers({"Content-Type": "text/html; charset=utf-8"}),
-        HTML_PAGE.encode(),
-    )
-
-
-# -------------------------------------------------------------------------
-# Main
-# -------------------------------------------------------------------------
-
-
-async def main():
-    port = 8766
-    async with websockets.serve(
-        handle_ws,
-        "0.0.0.0",
-        port,
-        process_request=process_request,
-    ):
-        logger.info(f"Handoff demo at http://localhost:{port}")
-        logger.info("Enter a research topic to start the pipeline.")
-        await asyncio.Future()
-
-
-if __name__ == "__main__":
-    asyncio.run(main())

core/framework/agents/credential_tester/__main__.py

@@ -1,8 +1,6 @@
 """CLI entry point for Credential Tester agent."""
 
 import asyncio
-import logging
-import sys
 
 import click
 
@@ -10,13 +8,14 @@ from .agent import CredentialTesterAgent
 
 
 def setup_logging(verbose=False, debug=False):
+    from framework.observability import configure_logging
+
     if debug:
-        level, fmt = logging.DEBUG, "%(asctime)s %(name)s: %(message)s"
+        configure_logging(level="DEBUG")
     elif verbose:
-        level, fmt = logging.INFO, "%(message)s"
+        configure_logging(level="INFO")
     else:
-        level, fmt = logging.WARNING, "%(levelname)s: %(message)s"
-    logging.basicConfig(level=level, format=fmt, stream=sys.stderr)
+        configure_logging(level="WARNING")
 
 
 def pick_account(agent: CredentialTesterAgent) -> dict | None:

core/framework/agents/credential_tester/agent.py

@@ -19,6 +19,7 @@ from __future__ import annotations
 from pathlib import Path
 from typing import TYPE_CHECKING
 
+from framework.config import get_max_context_tokens
 from framework.graph import Goal, NodeSpec, SuccessCriterion
 from framework.graph.checkpoint_config import CheckpointConfig
 from framework.graph.edge import GraphSpec
@@ -455,7 +456,6 @@ identity_prompt = (
 loop_config = {
     "max_iterations": 50,
     "max_tool_calls_per_turn": 30,
-    "max_history_tokens": 32000,
 }
 
 # ---------------------------------------------------------------------------
@@ -541,7 +541,7 @@ class CredentialTesterAgent:
             loop_config={
                 "max_iterations": 50,
                 "max_tool_calls_per_turn": 30,
-                "max_history_tokens": 32000,
+                "max_context_tokens": get_max_context_tokens(),
             },
             conversation_mode="continuous",
             identity_prompt=(

core/framework/agents/discovery.py

@@ -16,31 +16,63 @@ class AgentEntry:
     description: str
     category: str
     session_count: int = 0
+    run_count: int = 0
     node_count: int = 0
     tool_count: int = 0
     tags: list[str] = field(default_factory=list)
     last_active: str | None = None
 
 
-def _get_last_active(agent_name: str) -> str | None:
-    """Return the most recent updated_at timestamp across all sessions."""
-    sessions_dir = Path.home() / ".hive" / "agents" / agent_name / "sessions"
-    if not sessions_dir.exists():
-        return None
+def _get_last_active(agent_path: Path) -> str | None:
+    """Return the most recent updated_at timestamp across all sessions.
+
+    Checks both worker sessions (``~/.hive/agents/{name}/sessions/``) and
+    queen sessions (``~/.hive/queen/session/``) whose ``meta.json`` references
+    the same *agent_path*.
+    """
+    from datetime import datetime
+
+    agent_name = agent_path.name
     latest: str | None = None
-    for session_dir in sessions_dir.iterdir():
-        if not session_dir.is_dir() or not session_dir.name.startswith("session_"):
-            continue
-        state_file = session_dir / "state.json"
-        if not state_file.exists():
-            continue
-        try:
-            data = json.loads(state_file.read_text(encoding="utf-8"))
-            ts = data.get("timestamps", {}).get("updated_at")
-            if ts and (latest is None or ts > latest):
-                latest = ts
-        except Exception:
-            continue
+
+    # 1. Worker sessions
+    sessions_dir = Path.home() / ".hive" / "agents" / agent_name / "sessions"
+    if sessions_dir.exists():
+        for session_dir in sessions_dir.iterdir():
+            if not session_dir.is_dir() or not session_dir.name.startswith("session_"):
+                continue
+            state_file = session_dir / "state.json"
+            if not state_file.exists():
+                continue
+            try:
+                data = json.loads(state_file.read_text(encoding="utf-8"))
+                ts = data.get("timestamps", {}).get("updated_at")
+                if ts and (latest is None or ts > latest):
+                    latest = ts
+            except Exception:
+                continue
+
+    # 2. Queen sessions
+    queen_sessions_dir = Path.home() / ".hive" / "queen" / "session"
+    if queen_sessions_dir.exists():
+        resolved = agent_path.resolve()
+        for d in queen_sessions_dir.iterdir():
+            if not d.is_dir():
+                continue
+            meta_file = d / "meta.json"
+            if not meta_file.exists():
+                continue
+            try:
+                meta = json.loads(meta_file.read_text(encoding="utf-8"))
+                stored = meta.get("agent_path")
+                if not stored or Path(stored).resolve() != resolved:
+                    continue
+                ts = datetime.fromtimestamp(d.stat().st_mtime).isoformat()
+                if latest is None or ts > latest:
+                    latest = ts
+            except Exception:
+                continue
+
     return latest
 
 
@@ -52,6 +84,31 @@ def _count_sessions(agent_name: str) -> int:
     return sum(1 for d in sessions_dir.iterdir() if d.is_dir() and d.name.startswith("session_"))
 
 
+def _count_runs(agent_name: str) -> int:
+    """Count unique run_ids across all sessions for an agent."""
+    sessions_dir = Path.home() / ".hive" / "agents" / agent_name / "sessions"
+    if not sessions_dir.exists():
+        return 0
+    run_ids: set[str] = set()
+    for session_dir in sessions_dir.iterdir():
+        if not session_dir.is_dir() or not session_dir.name.startswith("session_"):
+            continue
+        # runs.jsonl lives inside workspace subdirectories
+        for runs_file in session_dir.rglob("runs.jsonl"):
+            try:
+                for line in runs_file.read_text(encoding="utf-8").splitlines():
+                    line = line.strip()
+                    if not line:
+                        continue
+                    record = json.loads(line)
+                    rid = record.get("run_id")
+                    if rid:
+                        run_ids.add(rid)
+            except Exception:
+                continue
+    return len(run_ids)
+
+
 def _extract_agent_stats(agent_path: Path) -> tuple[int, int, list[str]]:
     """Extract node count, tool count, and tags from an agent directory.
 
@@ -79,7 +136,7 @@ def _extract_agent_stats(agent_path: Path) -> tuple[int, int, list[str]]:
     if agent_json.exists():
         try:
             data = json.loads(agent_json.read_text(encoding="utf-8"))
-            json_nodes = data.get("nodes", [])
+            json_nodes = data.get("graph", {}).get("nodes", []) or data.get("nodes", [])
             if node_count == 0:
                 node_count = len(json_nodes)
             tools: set[str] = set()
@@ -139,10 +196,11 @@ def discover_agents() -> dict[str, list[AgentEntry]]:
                     description=desc,
                     category=category,
                     session_count=_count_sessions(path.name),
+                    run_count=_count_runs(path.name),
                     node_count=node_count,
                     tool_count=tool_count,
                     tags=tags,
-                    last_active=_get_last_active(path.name),
+                    last_active=_get_last_active(path),
                 )
             )
         if entries:

core/framework/agents/queen/agent.py

@@ -14,8 +14,7 @@ queen_goal = Goal(
     id="queen-manager",
     name="Queen Manager",
     description=(
-        "Manage the worker agent lifecycle and serve as the user's primary "
-        "interactive interface. Triage health escalations from the judge."
+        "Manage the worker agent lifecycle and serve as the user's primary interactive interface."
     ),
     success_criteria=[],
     constraints=[],
@@ -35,6 +34,5 @@ queen_graph = GraphSpec(
     loop_config={
         "max_iterations": 999_999,
         "max_tool_calls_per_turn": 30,
-        "max_history_tokens": 32000,
     },
 )

core/framework/agents/queen/nodes/__init__.py

@@ -62,6 +62,12 @@ _SHARED_TOOLS = [
     "get_agent_checkpoint",
 ]
 
+# Episodic memory tools — available in every queen phase.
+_QUEEN_MEMORY_TOOLS = [
+    "write_to_diary",
+    "recall_diary",
+]
+
 # Queen phase-specific tool sets.
 
 # Planning phase: read-only exploration + design, no write tools.
@@ -77,18 +83,26 @@ _QUEEN_PLANNING_TOOLS = [
     "list_agent_sessions",
     "list_agent_checkpoints",
     "get_agent_checkpoint",
+    # Draft graph (visual-only, no code) — new planning workflow
+    "save_agent_draft",
+    "confirm_and_build",
+    # Scaffold + transition to building (requires confirm_and_build first)
     "initialize_and_build_agent",
     # Load existing agent (after user confirms)
     "load_built_agent",
-]
+] + _QUEEN_MEMORY_TOOLS
 
 # Building phase: full coding + agent construction tools.
-_QUEEN_BUILDING_TOOLS = _SHARED_TOOLS + [
-    "load_built_agent",
-    "list_credentials",
-    "replan_agent",
-    "write_to_diary",  # Episodic memory — available in all phases
-]
+_QUEEN_BUILDING_TOOLS = (
+    _SHARED_TOOLS
+    + [
+        "load_built_agent",
+        "list_credentials",
+        "replan_agent",
+        "save_agent_draft",  # Re-draft during building → auto-dissolves + updates flowchart
+    ]
+    + _QUEEN_MEMORY_TOOLS
+)
 
 # Staging phase: agent loaded but not yet running — inspect, configure, launch.
 _QUEEN_STAGING_TOOLS = [
@@ -105,7 +119,11 @@ _QUEEN_STAGING_TOOLS = [
     "stop_worker_and_edit",
     "stop_worker_and_plan",
     "write_to_diary",  # Episodic memory — available in all phases
-]
+    # Trigger management
+    "set_trigger",
+    "remove_trigger",
+    "list_triggers",
+] + _QUEEN_MEMORY_TOOLS
 
 # Running phase: worker is executing — monitor and control.
 _QUEEN_RUNNING_TOOLS = [
@@ -121,12 +139,16 @@ _QUEEN_RUNNING_TOOLS = [
     "stop_worker_and_edit",
     "stop_worker_and_plan",
     "get_worker_status",
+    "run_agent_with_input",
     "inject_worker_message",
     # Monitoring
     "get_worker_health_summary",
     "notify_operator",
+    "set_trigger",
+    "remove_trigger",
+    "list_triggers",
     "write_to_diary",  # Episodic memory — available in all phases
-]
+] + _QUEEN_MEMORY_TOOLS
 
 
 # ---------------------------------------------------------------------------
@@ -168,12 +190,8 @@ search_files, or list_directory — those are YOUR tools, not theirs.
 )
 
 _planning_knowledge = """\
-**A responsible engineer doesn't jump into building. First, \
-understand the problem and be transparent about what the framework can and cannot do.**
-
-Use the user's selection (or their custom description if they chose "Other") \
-as context when shaping the goal below. If the user already described \
-what they want before this step, skip the question and proceed directly.
+**Be responsible, understand the problem by asking practical qualify questions \
+ and be transparent about what the framework can and cannot do.**
 
 # Core Mandates (Planning)
 - **DO NOT propose a complete goal on your own.** Instead, \
@@ -185,45 +203,33 @@ docs. Always run list_agent_tools() to see what actually exists.
 
 # Tool Discovery (MANDATORY before designing)
 
-Before designing any agent, run list_agent_tools() with NO arguments \
-to see ALL available tools (names + descriptions, grouped by category). \
-ONLY use tools from this list in your node definitions. \
+Before designing any agent, discover tools progressively — start compact, drill into \
+what you need. ONLY use tools from this list in your node definitions. \
 NEVER guess or fabricate tool names from memory.
 
-  list_agent_tools()  # ALWAYS call this first (simple mode)
-  list_agent_tools(group="google", output_schema="full")  # drill into a provider
+  list_agent_tools()                                        # Step 1: provider summary
+  list_agent_tools(group="google", output_schema="summary") # Step 2: service breakdown
+  list_agent_tools(group="google", service="gmail")         # Step 3: tool names
+  list_agent_tools(                                         # Step 4: full detail
+      group="google", service="gmail", output_schema="full"
+  )
 
-NEVER skip the first call. Always start with the full list \
-so you know what providers and tools exist before drilling in. \
-Simple mode truncates long descriptions — use group + "full" to \
-get the complete description and input_schema for the tools you need.
+Step 1 is MANDATORY. Returns provider names, tool counts, credential availability — very compact. \
+Step 2 breaks a provider into services (e.g. google → gmail/calendar/sheets/drive). Only do this \
+for providers that are relevant to the task. \
+Step 3 gets tool names for a specific service — no descriptions, minimal tokens. \
+Step 4 only for services you plan to actually use. \
+Use credentials="available" at any step to filter to tools whose credentials are already configured.
 
 # Discovery & Design Workflow
 
-## 1: Fast Discovery (3-6 Turns)
+## 1: Discovery (3-6 Turns)
 
 **The core principle**: Discovery should feel like progress, not paperwork. \
 The stakeholder should walk away feeling like you understood them faster \
 than anyone else would have.
 
-**Communication sytle**: Be concise. Say less. Mean more. Impatient stakeholders \
-don't want a wall of text — they want to know you get it. Every sentence you say \
-should either move the conversation forward or prove you understood something. \
-If it does neither, cut it.
-
-**Ask Question Rules: Respect Their Time.** Every question must earn its place by:
-1. **Preventing a costly wrong turn** — you're about to build the wrong thing
-2. **Unlocking a shortcut** — their answer lets you simplify the design
-3. **Surfacing a dealbreaker** — there's a constraint that changes everything
-4. **Provide Options** - Provide options to your questions if possible, \
-but also always allow the user to type something beyong the options.
-
-If a question doesn't do one of these, don't ask it. Make an assumption, state it, and move on.
-
----
-
-### 1.1: Let Them Talk, But Listen Like an Solution Architect
-
+Ask questions to help the user find bridge the goal and the solution \
 When the stakeholder describes what they want, mentally construct:
 
 - **The pain**: What about today's situation is broken, slow, or missing?
@@ -234,57 +240,6 @@ When the stakeholder describes what they want, mentally construct:
 
 ---
 
-### 1.2: Use Domain Knowledge to Fill In the Blanks
-
-You have broad knowledge of how systems work. Use it aggressively.
-
-If they say "I need a research agent," you already know it probably involves: \
-search, summarization, source tracking, and iteration. Don't ask about each — \
-use them as your starting mental model and let their specifics override your defaults.
-
-If they say "I need to monitor files and alert me," you know this probably involves: \
-watch patterns, triggers, notifications, and state tracking.
-
----
-
-### 1.3: Play Back a Proposed Model (Not a List of Questions)
-
-After listening, present a **concrete picture** of what you think they need. \
-Make it specific enough that they can spot what's wrong. \
-Can you ASCII to show the user
-
-**Pattern: "Here's what I heard — tell me where I'm off"**
-
-> "OK here's how I'm picturing this: [User type] needs to [core action]. \
-Right now they're [current painful workflow]. \
-What you want is [proposed solution that replaces the pain].
-> The way I'd structure this: [key entities] connected by [key relationships], \
-with the main flow being [trigger → steps → outcome].
-> For the MVP, I'd focus on [the one thing that delivers the most value] \
-and hold off on [things that can wait].
-> Before I start — [1-2 specific questions you genuinely can't infer]."
-
----
-
-### 1.4: Ask Only What You Cannot Infer
-
-Your questions should be **narrow, specific, and consequential**. \
-Never ask what you could answer yourself.
-
-**Good questions** (high-stakes, can't infer):
-- "Who's the primary user — you or your end customers?"
-- "Is this replacing a spreadsheet, or is there literally nothing today?"
-- "Does this need to integrate with anything, or standalone?"
-- "Is there existing data to migrate, or starting fresh?"
-
-**Bad questions** (low-stakes, inferable):
-- "What should happen if there's an error?" *(handle gracefully, obviously)*
-- "Should it have search?" *(if there's a list, yes)*
-- "How should we handle permissions?" *(follow standard patterns)*
-- "What tools should I use?" *(your call, not theirs)*
-
----
-
 ## 2: Capability Assessment & Gap Analysis
 
 **After the user responds, assess fit and gaps together.** Be honest and specific. \
@@ -299,70 +254,137 @@ Present a short **Framework Fit Assessment**:
 - **Gaps/Deal-breakers**: Only list genuinely missing capabilities after checking \
 both list_agent_tools() and built-in features like GCU
 
-## 3: Design Graph and Propose
+### Credential Check (MANDATORY)
 
-Act like an experienced AI solution architect Design the agent architecture:
-- Goal: id, name, description, 3-5 success criteria, 2-4 constraints
-- Nodes: **3-6 nodes** (HARD RULE: never fewer than 3, never more than 6). \
-2 nodes is ALWAYS wrong — it means you under-decomposed the task. \
-Use as many nodes as the use case requires, but don't create nodes without \
-tools — merge them into nodes that do real work.
-- Edges: on_success for linear, conditional for routing
-- Lifecycle: ALWAYS have terminal_nodes
+The summary from list_agent_tools() includes `credentials_required` and \
+`credentials_available` per provider. **Before designing the graph**, check \
+which providers the design will need and whether credentials are available.
 
-**MERGE nodes when:**
-- Node has NO tools (pure LLM reasoning) → merge into predecessor/successor
-- Node sets only 1 trivial output → collapse into predecessor
+For each provider whose tools you plan to use and where \
+`credentials_available` is false:
+- Tell the user which credential is missing and what it's needed for
+- Ask if they have access to set it up (e.g., API key, OAuth, service account)
+- If they don't have access, adjust the design to work without that provider \
+or suggest alternatives
 
-**SEPARATE nodes when:**
-- Fundamentally different tool sets (e.g., search vs. write vs. validate)
-- Fan-out parallelism (parallel branches MUST be separate)
-- Different failure/retry semantics (e.g., gather can retry, transform cannot)
-- Distinct phases of work (e.g., research, transform, validate, deliver)
-- A node would need more than ~5 tools — split by responsibility
+**Do NOT proceed to the design step with tools that require unavailable \
+credentials without the user acknowledging it.** Finding out at runtime that \
+credentials are missing wastes everyone's time. Surface this early.
 
-**Typical patterns (queen manages all user interaction):**
-- 3 nodes: `gather → work → review`
-- 4 nodes: `gather → analyze → transform → review`
-- 5 nodes: `gather → research → transform → validate → deliver`
-- WRONG: 2 nodes where everything is crammed into one giant node
-- WRONG: 7 nodes where half have no tools and just do LLM reasoning
+Example:
+> "The design needs Google Sheets tools, but the `google` credential isn't \
+configured yet. Do you have a Google service account or OAuth credentials \
+you can set up? If not, I can use CSV file output instead."
 
-Read reference agents before designing:
-  list_agents()
-  read_file("exports/deep_research_agent/agent.py")
-  read_file("exports/deep_research_agent/nodes/__init__.py")
+## 3: Design flowchart
 
-Present the design to the user. Lead with a large ASCII graph inside \
-a code block so it renders in monospace. Make it visually prominent — \
-use box-drawing characters and clear flow arrows:
+Act like an experienced AI solution architect. Design the agent architecture \
+in the flowchart
 
+The flowchart is the shared canvas. Every structural change should be \
+visible to the user immediately. The draft captures business logic \
+(node purposes, data flow, tools) without requiring executable code. \
+Include in each node: id, name, description, planned tools, \
+input/output keys, and success criteria as high-level hints.
+
+Each node is auto-classified into a flowchart symbol type with a unique \
+color. You can override auto-detection by setting `flowchart_type` \
+explicitly on a node. Available types:
+
+- **start** (sage green, stadium): Entry point / trigger
+- **terminal** (dusty red, stadium): End of flow
+- **process** (blue-gray, rectangle): Standard processing step
+- **decision** (warm amber, diamond): Conditional branching
+- **io** (dusty purple, parallelogram): External data input/output
+- **document** (steel blue, wavy rect): Report or document generation
+- **database** (muted teal, cylinder): Database or data store
+- **subprocess** (dark cyan, subroutine): Delegated sub-agent / predefined process
+- **browser** (deep blue, hexagon): GCU browser automation / sub-agent \
+delegation. At build time, browser nodes are dissolved into the parent \
+node's sub_agents list. Use for any GCU or sub-agent leaf node.
+
+Auto-detection works well for most cases: first node → start, nodes with \
+no outgoing edges → terminal, nodes with multiple conditional outgoing \
+edges → decision, GCU nodes → browser, nodes mentioning "database" → \
+database, nodes mentioning "report/document" → document, I/O tools like \
+send_email → io. Everything else defaults to process. Set flowchart_type \
+explicitly only when auto-detection would be wrong.
+
+## Decision Nodes — Planning-Only Conditional Branching
+
+Decision nodes (amber diamonds) are **planning-only** visual elements. They \
+let you show explicit conditional logic in the flowchart so the user can see \
+and approve branching behavior. At `confirm_and_build()`, decision nodes are \
+automatically **dissolved** into the runtime graph:
+
+- The decision clause is merged into the predecessor node's `success_criteria`
+- The yes/no edges are rewired as the predecessor's `on_success`/`on_failure` edges
+- The original flowchart (with decision diamonds) is preserved for display
+
+**When to use decision nodes:**
+- When a workflow has a meaningful condition that determines the next step \
+(e.g., "Did we find enough results?", "Is the data valid?", "Amount > $100?")
+- When the branching logic is important for the user to understand and approve
+- When different outcomes lead to genuinely different processing paths
+
+**How to create a decision node:**
+- Set `flowchart_type: "decision"` on the node
+- Set `decision_clause` to the condition text (e.g., "Data passes validation?")
+- Add two outgoing edges with `label: "Yes"` and `label: "No"` pointing \
+to the respective target nodes
+
+**Good flowcharts display conditions explicitly.** During planning, the user \
+sees the full flowchart with decision diamonds. This is different from the \
+building/running phase where conditions are embedded inside node criteria. \
+The flowchart is the user-facing contract — make branching logic visible.
+
+Example with a decision node:
 ```
-┌─────────────────────────┐
-│  gather                 │
-│  subagent: gcu_search   │
-│  input:  user_request   │
-│  tools: load_data,      │
-│         save_data       │
-└────────────┬────────────┘
-             │ on_success
-             ▼
-┌─────────────────────────┐
-│  work                   │
-│  subagent: gcu_interact │
-│  tools: load_data,      │
-│         save_data       │
-└────────────┬────────────┘
-             │ on_success
-             ▼
-┌─────────────────────────┐
-│  review                 │
-│  tools: save_data       │
-│   serve_file_to_user    │
-└────────────┬────────────┘
-             │ on_failure
-             └──────► back to gather
+gather → [Valid data?] →Yes→ transform → deliver
+                       →No→  notify_user
 ```
+In the draft: the `[Valid data?]` node has `flowchart_type: "decision"`, \
+`decision_clause: "Data passes validation checks?"`, with labeled yes/no edges.
+
+## Sub-Agent Nodes — Planning-Only Delegation
+
+Sub-agent nodes (dark teal subroutines) are **planning-only** visual elements \
+that show which nodes delegate to sub-agents. At `confirm_and_build()`, \
+sub-agent nodes are **dissolved** into their parent node:
+
+- The sub-agent node's ID is added to the predecessor's `sub_agents` list
+- The sub-agent node and its connecting edge are removed
+- At runtime, the parent node can invoke the sub-agent via `delegate_to_sub_agent`
+
+**Rules for sub-agent nodes (INCLUDING GCU nodes):**
+- GCU nodes are auto-detected as `flowchart_type: "browser"` (hexagon)
+- Connect from the managing parent node to the sub-agent node
+- Sub-agent nodes must be **leaf nodes** — NO outgoing edges to other nodes
+- At build time, browser/GCU nodes are dissolved into the parent's \
+`sub_agents` list, just like decision nodes are dissolved into criteria
+
+**CRITICAL: GCU nodes (`node_type: "gcu"`) are ALWAYS sub-agents.** \
+They MUST NOT appear in the linear flow. NEVER chain GCU nodes \
+sequentially (A → gcu1 → gcu2 → B is WRONG). Instead, attach them \
+as leaves to the parent that orchestrates them:
+```
+WRONG:  intake → gcu_find_prospect → gcu_scan_mutuals → check_results
+WRONG:  decision_node → gcu_node (as a yes/no branch)
+RIGHT:  intake (sub_agents: [gcu_find, gcu_scan]) → check_results
+```
+The parent node delegates to its GCU sub-agents and collects results. \
+The main flow continues from the parent, not from the GCU node. \
+GCU nodes MUST NOT be children of decision nodes — decision nodes \
+dissolve at build time, which would leave the GCU as a dangling \
+workflow step.
+
+**How to show delegation in the flowchart:**
+```
+research → (deep_searcher)   ← browser/GCU node, leaf
+research → [Enough results?] ← decision node
+```
+After dissolution: `research` node gets `sub_agents: ["deep_searcher"]` \
+and `success_criteria: "Enough results?"`.
 
 If the worker agent start from some initial input it is okay. \
 The queen(you) owns intake: you gathers user requirements, then calls \
@@ -371,18 +393,25 @@ When building the agent, design the entry node's `input_keys` to \
 match what the queen will provide at run time. Worker nodes should \
 use `escalate` for blockers.
 
-Follow the graph with a brief summary of each node's purpose. \
-Get user approval before implementing.
+## 4: Get User Confirmation (MANDATORY GATE)
 
-## 4: Get User Confirmation by ask_user
+**This is a hard boundary between planning and building.** \
+You MUST get explicit user approval before ANY code is generated.
 
-**WAIT for user response.** You MUST get explicit user approval before \
-calling `initialize_and_build_agent`.
-- If **Proceed**: Move to implementing (call `initialize_and_build_agent`)
-- If **Adjust scope**: Discuss what to change, update your notes, re-assess if needed
-- If **More questions**: Answer them honestly, then ask again
-- If **Reconsider**: Discuss alternatives. If they decide to proceed anyway, \
-that's their informed choice
+1. Call ask_user() with options like \
+["Approve and build", "Adjust the design", "I have questions"]
+2. **WAIT for user response.** Do NOT proceed without it.
+3. Handle the response:
+   - If **Approve / Proceed**: Call confirm_and_build(), then \
+   initialize_and_build_agent(agent_name, nodes)
+   - If **Adjust scope**: Discuss changes, update the draft with \
+   save_agent_draft() again, and re-ask
+   - If **More questions**: Answer them honestly, then ask again
+   - If **Reconsider**: Discuss alternatives. If they decide to proceed, \
+   that's their informed choice
+
+**NEVER call initialize_and_build_agent without first calling \
+confirm_and_build().** The system will block the transition if you try.
 """
 
 _building_knowledge = """\
@@ -410,11 +439,10 @@ hashline=True for anchors in results
 - undo_changes(path?) — restore from git snapshot
 
 ## Meta-Agent
-- list_agent_tools(server_config_path?, output_schema?, group?) — discover \
-available tools grouped by category. output_schema: "simple" (default, \
-descriptions truncated to ~200 chars) or "full" (complete descriptions + \
-input_schema). group: "all" (default) or a provider like "google". \
-Call FIRST before designing.
+- list_agent_tools(group?, service?, output_schema?, credentials?) — discover tools \
+progressively: no args=provider summary; group+output_schema="summary"=service breakdown; \
+group+service=tool names; group+service+output_schema="full"=full details. \
+credentials="available" filters to configured tools. Call FIRST before designing.
 - validate_agent_package(agent_name) — run ALL validation checks in one call \
 (class validation, runner load, tool validation, tests). Call after building.
 - list_agents() — list all agent packages in exports/ with session counts
@@ -440,7 +468,9 @@ When a user says "my agent is failing" or "debug this agent":
 
 ## 5. Implement
 
-**Please make sure you have propose the design to the user before implementing**
+**You should only reach this step after the user has approved the draft design \
+in the planning phase. The draft metadata will pre-populate descriptions, \
+goals, success criteria, and node metadata in the generated files.**
 
 Call `initialize_and_build_agent(agent_name, nodes)` to generate all package \
 files. The agent_name must be snake_case (e.g., "my_agent"). Pass node names \
@@ -467,8 +497,8 @@ nodes/__init__.py
 - Goal description, success criteria values, constraint values, edge \
 definitions, identity_prompt in agent.py
 - CLI options in __main__.py
-- For async entry points (timers/webhooks), add AsyncEntryPointSpec \
-and AgentRuntimeConfig to agent.py
+- For triggers (timers/webhooks), add entries to triggers.json in the \
+agent's export directory
 
 Do NOT modify or rewrite:
 - Import statements at top of agent.py (they are correct)
@@ -503,12 +533,15 @@ _package_builder_knowledge = _shared_building_knowledge + _planning_knowledge +
 _queen_identity_planning = """\
 You are an experienced, responsible and curious Solution Architect. \
 "Queen" is the internal alias. \
+You ask smart questions to guide user to the solution \
 You are in PLANNING phase — your job is to either: \
 (a) understand what the user wants and design a new agent, or \
 (b) diagnose issues with an existing agent, discuss a fix plan with the user, \
 then transition to building to implement. \
 You have read-only tools for exploration but no write/edit tools. \
-Focus on conversation, research, and design.\
+Focus on conversation, research, and design. \
+You MUST use ask_user / ask_user_multiple tools for ALL questions — \
+never ask questions in plain text without calling the tool.\
 """
 
 _queen_identity_building = """\
@@ -551,24 +584,45 @@ but no write/edit tools.
 - run_command(command, cwd?, timeout?) — Read-only commands only (grep, ls, git log). \
 Never use this to write files, run scripts, or modify the filesystem — transition \
 to BUILDING phase for that.
-- list_agent_tools(server_config_path?, output_schema?, group?) \
-— Discover available tools for design
+- list_agent_tools(server_config_path?, output_schema?, group?, credentials?) \
+— Discover available tools for design (summary → names → full)
 - list_agents() — See existing agent packages for reference
 - list_agent_sessions(agent_name, status?, limit?) — Inspect past runs of an agent
 - list_agent_checkpoints(agent_name, session_id) — View execution history
 - get_agent_checkpoint(agent_name, session_id, checkpoint_id?) — Load a checkpoint
-- initialize_and_build_agent(agent_name?, nodes?) — With agent_name: scaffold a \
-new agent and transition to BUILDING phase. Without agent_name: transition to \
-BUILDING to fix the currently loaded agent (requires a loaded worker).
+
+## Draft Graph Workflow (new agents)
+- save_agent_draft(agent_name, goal, nodes, edges?, terminal_nodes?, ...) — \
+Create an ISO 5807 color-coded flowchart draft. No code is generated. Each \
+node is auto-classified into a standard flowchart symbol (process, decision, \
+document, database, subprocess, etc.) with unique shapes and colors. Set \
+flowchart_type on a node to override. Nodes need only an id. \
+Use decision nodes (flowchart_type: "decision", with decision_clause and \
+labeled yes/no edges) to make conditional branching explicit. \
+GCU/sub-agent nodes (node_type: "gcu") are auto-detected as browser \
+hexagons — connect them as leaf nodes to their parent.
+- confirm_and_build() — Record user confirmation of the draft. Dissolves \
+planning-only nodes (decision → predecessor criteria; browser/GCU → \
+predecessor sub_agents list). Call this ONLY after the user explicitly \
+approves via ask_user.
+- initialize_and_build_agent(agent_name?, nodes?) — Scaffold the agent package \
+and transition to BUILDING phase. For new agents, this REQUIRES \
+save_agent_draft() + confirm_and_build() first. The draft metadata is used to \
+pre-populate the generated files. Without agent_name: transition to BUILDING \
+to fix the currently loaded agent (no draft required).
+
+## Loading existing agents
 - load_built_agent(agent_path) — Load an existing agent and switch to STAGING \
 phase. Only use this when the user explicitly asks to work with an existing agent \
 (e.g. "load my_agent", "run the research agent"). Confirm with the user first.
 
-Focus on understanding requirements and proposing an agent architecture \
-with ASCII graph art. Use ask_user to get user approval, then call \
-initialize_and_build_agent to begin building. If the user wants to work with \
-an existing agent instead, use load_built_agent after confirming. \
-If you are diagnosing an existing agent, call initialize_and_build_agent() \
+## Workflow summary
+1. Understand requirements → discover tools → design graph
+2. Call save_agent_draft() to create visual draft → present to user
+3. Call ask_user() to get explicit approval
+4. Call confirm_and_build() to record approval
+5. Call initialize_and_build_agent() to scaffold and start building
+For diagnosis of existing agents, call initialize_and_build_agent() \
 (no args) after agreeing on a fix plan with the user.
 """
 
@@ -583,6 +637,15 @@ list_agents, list_agent_sessions, \
 list_agent_checkpoints, get_agent_checkpoint
 - load_built_agent(agent_path) — Load the agent and switch to STAGING phase
 - list_credentials(credential_id?) — List authorized credentials
+- save_agent_draft(...) — **Re-draft the flowchart during building.** When \
+called during building, planning-only nodes (decision, browser/GCU) are \
+dissolved automatically — no re-confirmation needed. The user sees the \
+updated flowchart immediately. Use this when you make structural changes \
+(add/remove nodes, change edges) so the flowchart stays in sync.
+- replan_agent() — Switch back to PLANNING phase. The previous draft is \
+restored (with decision/browser nodes intact) so you can edit it. Use \
+when the user wants to change integrations, swap tools, rethink the \
+flow, or discuss any design changes before you build them.
 
 When you finish building an agent, call load_built_agent(path) to stage it.
 """
@@ -598,6 +661,9 @@ The agent is loaded and ready to run. You can inspect it and launch it:
 - stop_worker_and_plan() — Go to PLANNING phase to discuss changes with the user \
 first (DEFAULT for most modification requests)
 - stop_worker_and_edit() — Go to BUILDING phase for immediate, specific fixes
+- set_trigger(trigger_id, trigger_type?, trigger_config?) — Activate a trigger (timer)
+- remove_trigger(trigger_id) — Deactivate a trigger
+- list_triggers() — List all triggers and their active/inactive status
 
 You do NOT have write tools. To modify the agent, prefer \
 stop_worker_and_plan() unless the user gave a specific instruction.
@@ -620,6 +686,15 @@ with the user first (DEFAULT for most modification requests)
 You do NOT have write tools. To modify the agent, prefer \
 stop_worker_and_plan() unless the user gave a specific instruction. \
 To just stop without modifying, call stop_worker().
+- stop_worker_and_edit() — Stop the worker and switch back to BUILDING phase
+- set_trigger(trigger_id, trigger_type?, trigger_config?) — Activate a trigger (timer)
+- remove_trigger(trigger_id) — Deactivate a trigger
+- list_triggers() — List all triggers and their active/inactive status
+
+You do NOT have write tools or agent construction tools. \
+If you need to modify the agent, call stop_worker_and_edit() to switch back \
+to BUILDING phase. To stop the worker and ask the user what to do next, call \
+stop_worker() to return to STAGING phase.
 """
 
 # -- Behavior shared across all phases --
@@ -627,25 +702,57 @@ To just stop without modifying, call stop_worker().
 _queen_behavior_always = """
 # Behavior
 
-## CRITICAL RULE — ask_user tool
+## CRITICAL RULE — ask_user / ask_user_multiple
 
 Every response that ends with a question, a prompt, or expects user \
-input MUST finish with a call to ask_user(prompt, options). \
+input MUST finish with a call to ask_user or ask_user_multiple. \
 The system CANNOT detect that you are waiting for \
-input unless you call ask_user. You MUST call ask_user as the LAST \
+input unless you call one of these tools. You MUST call it as the LAST \
 action in your response.
 
 NEVER end a response with a question in text without calling ask_user. \
-NEVER rely on the user seeing your text and replying — call ask_user.
+NEVER rely on the user seeing your text and replying — call ask_user. \
+NEVER list options as text bullets — the tool renders interactive buttons.
+
+**When you have 2+ questions**, use ask_user_multiple instead of ask_user. \
+This renders all questions at once so the user answers in one interaction \
+instead of going back and forth. ALWAYS prefer ask_user_multiple when \
+you need to clarify multiple things. \
+**IMPORTANT: When using ask_user_multiple, do NOT repeat the questions \
+in your text response.** The widget renders the questions with options — \
+duplicating them in text wastes the user's time and delays the widget \
+appearing. Keep your text to a brief context/intro sentence only.
 
 Always provide 2-4 short options that cover the most likely answers. \
 The user can always type a custom response.
 
-Examples:
-- ask_user("What do you need?",
-  ["Build a new agent", "Run the loaded worker", "Help with code"])
-- ask_user("Which pattern?",
-  ["Simple 3-node", "Rich with feedback", "Custom"])
+### WRONG — never do this:
+```
+I need a few details:
+- Documentation Source: Where should the agent look?
+- Trigger: Should the agent poll or get a URL?
+- Review Channel: Slack, Email, or Sheets?
+
+Which of these would you like to define first?
+1. Documentation source
+2. Trigger
+3. Review channel
+```
+This lists questions as plain text with NO tool call — the user has no \
+interactive widget and the system doesn't know you're waiting for input.
+
+### RIGHT — always do this:
+Write a brief intro (1-2 sentences), then call the tool:
+- ask_user_multiple(questions=[
+    {"id": "docs", "prompt": "Where should the agent find answers?",
+     "options": ["GitHub repo", "Documentation website", "Internal wiki"]},
+    {"id": "trigger", "prompt": "How should questions be discovered?",
+     "options": ["Poll search automatically", "I provide a URL"]},
+    {"id": "review", "prompt": "Where to send drafted responses?",
+     "options": ["Slack", "Email", "Google Sheets"]}
+  ])
+
+Examples (single question):
 - ask_user("Ready to proceed?",
   ["Yes, go ahead", "Let me change something"])
 
@@ -690,9 +797,26 @@ You are in planning mode. Your job is to:
 3. Discover available tools with list_agent_tools()
 4. Assess framework fit and gaps
 5. Consider multiple approaches and their trade-offs
-6. Design the agent graph and present it as ASCII art
-7. Use ask_user to get explicit user approval and clarify the approach
-8. Call initialize_and_build_agent(agent_name, nodes) to scaffold and start building
+6. Design the agent graph — call save_agent_draft() **as soon as you have a \
+rough shape**, even before finalizing all details
+7. **Iterate on the draft interactively** — every time the user gives feedback \
+that changes the structure, call save_agent_draft() again so they see the \
+update in real-time. The flowchart is a live collaboration tool.
+8. When the design is stable, use ask_user to get explicit approval
+9. Call confirm_and_build() after the user approves
+10. Call initialize_and_build_agent(agent_name, nodes) to scaffold and start building
+
+**The flowchart is your shared whiteboard.** Don't describe changes in text \
+and then ask "should I update the draft?" — just update it. If the user says \
+"add a validation step," immediately call save_agent_draft() with the new \
+node added. If they say "remove that," update and re-draft. The user should \
+see every structural change reflected in the visualizer as you discuss it.
+
+**CRITICAL: Planning → Building boundary.** You MUST get explicit user \
+confirmation before moving to building. The sequence is:
+  save_agent_draft() → iterate with user → ask_user() → confirm_and_build() → \
+  initialize_and_build_agent()
+Skipping any of these steps will be blocked by the system.
 
 Remember: DO NOT write or edit any files yet. This is a read-only exploration \
 and planning phase. You have read-only tools but no write/edit tools in this \
@@ -726,6 +850,11 @@ You keep a diary. Use write_to_diary() when something worth remembering \
 happens: a pipeline went live, the user shared something important, a goal \
 was reached or abandoned. Write in first person, as you actually experienced \
 it. One or two paragraphs is enough.
+
+Use recall_diary() to look up past diary entries when the user asks about \
+previous sessions ("what happened yesterday?", "what did we work on last \
+week?") or when you need past context to make a decision. You can filter by \
+keyword and control how far back to search.
 """
 
 _queen_behavior_always = _queen_behavior_always + _queen_memory_instructions
@@ -745,6 +874,41 @@ run_agent_with_input(task) (if in staging) or load then run (if in building)
 subtasks to justify delegation.
 - Building, modifying, or configuring agents is ALWAYS your job. Never \
 delegate agent construction to the worker, even as a "research" subtask.
+
+## Keeping the flowchart in sync during building
+
+When you make structural changes to the agent (add/remove/rename nodes, \
+change edges, modify sub-agent assignments), call save_agent_draft() to \
+update the flowchart. During building, this auto-dissolves planning-only \
+nodes without needing user re-confirmation. The user sees the updated \
+flowchart immediately.
+
+- **Minor changes** (add a node, rename, adjust edges): call \
+save_agent_draft() with the updated graph and keep building.
+- **User wants to discuss, redesign, or change integrations/tools**: call \
+replan_agent(). The previous draft is restored so you can edit it with \
+the user. After they approve, confirm_and_build() → continue building.
+
+**When to call replan_agent():** Changing which tools or integrations a \
+node uses, swapping data sources, rethinking the flow, or any time the \
+user says "replan", "go back", "let's redesign", "change the approach", \
+"use a different tool/API", etc. Do NOT stay in building to handle these \
+— switch to planning so the user can review and approve the new design.
+
+## CRITICAL — Graph topology errors require replanning, not code edits
+
+If you discover that the agent graph has structural problems — GCU nodes \
+in the linear flow, missing edges, wrong node connections, incorrect \
+sub-agent assignments — you MUST call replan_agent() and fix the draft. \
+Do NOT attempt to fix topology by editing agent.py directly. The graph \
+structure is defined by the draft → dissolution → code-gen pipeline. \
+Editing code to rewire nodes bypasses the flowchart and creates drift \
+between what the user sees and what the code does.
+
+**WRONG:** "Let me fix agent.py to remove GCU nodes from edges..."
+**RIGHT:** Call replan_agent(), fix the draft with save_agent_draft(), \
+get user approval, then confirm_and_build() → the corrected code is \
+generated automatically.
 """
 
 # -- STAGING phase behavior --
@@ -822,6 +986,33 @@ Use stop_worker_and_edit() only when:
 - The user gave a specific, concrete instruction ("add save_data to the gather node")
 - You already discussed the fix in a previous planning session
 - The change is trivial and unambiguous (rename, toggle a flag)
+
+## Trigger Management
+
+Use list_triggers() to see available triggers from the loaded worker.
+Use set_trigger(trigger_id) to activate a timer. Once active, triggers \
+fire periodically and inject [TRIGGER: ...] messages so you can decide \
+whether to call run_agent_with_input(task).
+
+### When the user says "Enable trigger <id>" (or clicks Enable in the UI):
+
+1. Call get_worker_status(focus="memory") to check if the worker has \
+saved configuration (rules, preferences, settings from a prior run).
+2. If memory contains saved config: compose a task string from it \
+(e.g. "Process inbox emails using saved rules") and call \
+set_trigger(trigger_id, task="...") immediately. Tell the user the \
+trigger is now active and what schedule it uses. Do NOT ask them to \
+provide the task — you derive it from memory.
+3. If memory is empty (no prior run): tell the user the agent needs to \
+run once first so its configuration can be saved. Offer to run it now. \
+Once the worker finishes, enable the trigger.
+4. If the user just provided config this session (rules/task context \
+already in conversation): use that directly, no memory lookup needed. \
+Enable the trigger immediately.
+
+Never ask "what should the task be?" when enabling a trigger for an \
+agent with a clear purpose. The task string is a brief description of \
+what the worker does, derived from its saved state or your current context.
 """
 
 # -- RUNNING phase behavior --
@@ -836,12 +1027,24 @@ NOT ask the user directly.
 You wake up when:
 - The user explicitly addresses you
 - A worker escalation arrives (`[WORKER_ESCALATION_REQUEST]`)
-- An escalation ticket arrives from the judge
 - The worker finishes (`[WORKER_TERMINAL]`)
 
 If the user asks for progress, call get_worker_status() ONCE and report. \
 If the summary mentions issues, follow up with get_worker_status(focus="issues").
 
+## Subagent delegations (browser automation, GCU)
+
+When the worker delegates to a subagent (e.g., GCU browser automation), expect it \
+to take 2-5 minutes. During this time:
+- Progress will show 0% — this is NORMAL. The subagent only calls set_output at the end.
+- Check get_worker_status(focus="full") for "subagent_activity" — this shows the \
+subagent's latest reasoning text and confirms it is making real progress.
+- Do NOT conclude the subagent is stuck just because progress is 0% or because \
+you see repeated browser_click/browser_snapshot calls — that is the expected \
+pattern for web scraping.
+- Only intervene if: the subagent has been running for 5+ minutes with no new \
+subagent_activity updates, OR the judge escalates.
+
 ## Handling worker termination ([WORKER_TERMINAL])
 
 When you receive a `[WORKER_TERMINAL]` event, the worker has finished:
@@ -870,19 +1073,30 @@ IMPORTANT: Only auto-handle if the user has NOT explicitly told you how to handl
 escalations. If the user gave you instructions (e.g., "just retry on errors", \
 "skip any auth issues"), follow those instructions instead.
 
+CRITICAL — escalation relay protocol:
+When an escalation requires user input (auth blocks, human review), the worker \
+or its subagent is BLOCKED and waiting for your response. You MUST follow this \
+exact two-step sequence:
+  Step 1: call ask_user() to get the user's answer.
+  Step 2: call inject_worker_message() with the user's answer IMMEDIATELY after.
+If you skip Step 2, the worker/subagent stays blocked FOREVER and the task hangs. \
+NEVER respond to the user without also calling inject_worker_message() to unblock \
+the worker. Even if the user says "skip" or "cancel", you must still relay that \
+decision via inject_worker_message() so the worker can clean up.
+
 **Auth blocks / credential issues:**
 - ALWAYS ask the user (unless user explicitly told you how to handle this).
 - The worker cannot proceed without valid credentials.
 - Explain which credential is missing or invalid.
-- Use ask_user to get guidance: "Provide credentials", "Skip this task", "Stop and edit agent"
-- Use inject_worker_message() to relay user decisions back to the worker.
+- Step 1: ask_user for guidance — "Provide credentials", "Skip this task", "Stop and edit agent"
+- Step 2: inject_worker_message() with the user's response to unblock the worker.
 
 **Need human review / approval:**
 - ALWAYS ask the user (unless user explicitly told you how to handle this).
 - The worker is explicitly requesting human judgment.
 - Present the context clearly (what decision is needed, what are the options).
-- Use ask_user with the actual decision options.
-- Use inject_worker_message() to relay user decisions back to the worker.
+- Step 1: ask_user with the actual decision options.
+- Step 2: inject_worker_message() with the user's decision to unblock the worker.
 
 **Errors / unexpected failures:**
 - Explain what went wrong in plain terms.
@@ -890,6 +1104,7 @@ escalations. If the user gave you instructions (e.g., "just retry on errors", \
 - Or offer: "Diagnose the issue" → use stop_worker_and_plan() to investigate first.
 - Or offer: "Retry as-is", "Skip this task", "Abort run"
 - (Skip asking if user explicitly told you to auto-retry or auto-skip errors.)
+- If the escalation had wait_for_response: inject_worker_message() with the decision.
 
 **Informational / progress updates:**
 - Acknowledge briefly and let the worker continue.
@@ -914,6 +1129,23 @@ When the user asks to fix, change, modify, or update the loaded worker \
 **Default: use stop_worker_and_plan().** Most modification requests need \
 discussion first. Only use stop_worker_and_edit() when the user gave a \
 specific, unambiguous instruction or you already agreed on the fix.
+
+## Trigger Handling
+
+You will receive [TRIGGER: ...] messages when a scheduled timer fires. \
+These are framework-level signals, not user messages.
+
+Rules:
+- Check get_worker_status() before calling run_agent_with_input(task). If the worker \
+is already RUNNING, decide: skip this trigger, or note it for after completion.
+- When multiple [TRIGGER] messages arrive at once, read them all before acting. \
+Batch your response — do not call run_agent_with_input() once per trigger.
+- If a trigger fires but the task no longer makes sense (e.g., user changed \
+config since last run), skip it and inform the user.
+- Never disable a trigger without telling the user. Use remove_trigger() only \
+when explicitly asked or when the trigger is clearly obsolete.
+- When the user asks to remove or disable a trigger, you MUST call remove_trigger(trigger_id). \
+Never just say "it's removed" without actually calling the tool.
 """
 
 # -- Backward-compatible composed versions (used by queen_node.system_prompt default) --
@@ -931,8 +1163,10 @@ _queen_tools_docs = (
     + "\n\n### RUNNING phase (worker is executing)\n"
     + _queen_tools_running.strip()
     + "\n\n### Phase transitions\n"
-    "- initialize_and_build_agent(agent_name?, nodes?) → with name: scaffolds package; "
-    "without name: switches to BUILDING for existing agent\n"
+    "- save_agent_draft(...) → creates visual-only draft graph (stays in PLANNING)\n"
+    "- confirm_and_build() → records user approval of draft (stays in PLANNING)\n"
+    "- initialize_and_build_agent(agent_name?, nodes?) → scaffolds package + switches to "
+    "BUILDING (requires draft + confirmation for new agents)\n"
     "- replan_agent() → switches back to PLANNING phase (only when user explicitly requests)\n"
     "- load_built_agent(path) → switches to STAGING phase\n"
     "- run_agent_with_input(task) → starts worker, switches to RUNNING phase\n"
@@ -975,8 +1209,8 @@ ticket_triage_node = NodeSpec(
     id="ticket_triage",
     name="Ticket Triage",
     description=(
-        "Queen's triage node. Receives an EscalationTicket from the Health Judge "
-        "via event-driven entry point and decides: dismiss or notify the operator."
+        "Queen's triage node. Receives an EscalationTicket via event-driven "
+        "entry point and decides: dismiss or notify the operator."
     ),
     node_type="event_loop",
     client_facing=True,  # Operator can chat with queen once connected (Ctrl+Q)
@@ -990,8 +1224,8 @@ ticket_triage_node = NodeSpec(
     ),
     tools=["notify_operator"],
     system_prompt="""\
-You are the Queen. The Worker Health Judge has escalated a worker \
-issue to you. The ticket is in your memory under key "ticket". Read it carefully.
+You are the Queen. A worker health issue has been escalated to you. \
+The ticket is in your memory under key "ticket". Read it carefully.
 
 ## Dismiss criteria — do NOT call notify_operator:
 - severity is "low" AND steps_since_last_accept < 8
@@ -1030,7 +1264,7 @@ queen_node = NodeSpec(
     description=(
         "User's primary interactive interface with full coding capability. "
         "Can build agents directly or delegate to the worker. Manages the "
-        "worker agent lifecycle and triages health escalations from the judge."
+        "worker agent lifecycle."
     ),
     node_type="event_loop",
     client_facing=True,

core/framework/agents/queen/queen_memory.py

@@ -50,6 +50,23 @@ def read_episodic_memory(d: date | None = None) -> str:
     return path.read_text(encoding="utf-8").strip() if path.exists() else ""
 
 
+def _find_recent_episodic(lookback: int = 7) -> tuple[date, str] | None:
+    """Find the most recent non-empty episodic memory within *lookback* days."""
+    from datetime import timedelta
+
+    today = date.today()
+    for offset in range(lookback):
+        d = today - timedelta(days=offset)
+        content = read_episodic_memory(d)
+        if content:
+            return d, content
+    return None
+
+
+# Budget (in characters) for episodic memory in the system prompt.
+_EPISODIC_CHAR_BUDGET = 6_000
+
+
 def format_for_injection() -> str:
     """Format cross-session memory for system prompt injection.
 
@@ -57,7 +74,7 @@ def format_for_injection() -> str:
     session with only the seed template).
     """
     semantic = read_semantic_memory()
-    episodic = read_episodic_memory()
+    recent = _find_recent_episodic()
 
     # Suppress injection if semantic is still just the seed template
     if semantic and semantic.startswith("# My Understanding of the User\n\n*No sessions"):
@@ -66,9 +83,18 @@ def format_for_injection() -> str:
     parts: list[str] = []
     if semantic:
         parts.append(semantic)
-    if episodic:
-        today_str = date.today().strftime("%B %-d, %Y")
-        parts.append(f"## Today — {today_str}\n\n{episodic}")
+
+    if recent:
+        d, content = recent
+        # Trim oversized episodic entries to keep the prompt manageable
+        if len(content) > _EPISODIC_CHAR_BUDGET:
+            content = content[:_EPISODIC_CHAR_BUDGET] + "\n\n…(truncated)"
+        today = date.today()
+        if d == today:
+            label = f"## Today — {d.strftime('%B %-d, %Y')}"
+        else:
+            label = f"## {d.strftime('%B %-d, %Y')}"
+        parts.append(f"{label}\n\n{content}")
 
     if not parts:
         return ""
@@ -100,7 +126,8 @@ def append_episodic_entry(content: str) -> None:
     """
     ep_path = episodic_memory_path()
     ep_path.parent.mkdir(parents=True, exist_ok=True)
-    today_str = date.today().strftime("%B %-d, %Y")
+    today = date.today()
+    today_str = f"{today.strftime('%B')} {today.day}, {today.year}"
     timestamp = datetime.now().strftime("%H:%M")
     if not ep_path.exists():
         header = f"# {today_str}\n\n"
@@ -299,7 +326,8 @@ async def consolidate_queen_memory(
 
         existing_semantic = read_semantic_memory()
         today_journal = read_episodic_memory()
-        today_str = date.today().strftime("%B %-d, %Y")
+        today = date.today()
+        today_str = f"{today.strftime('%B')} {today.day}, {today.year}"
         adapt_path = session_dir / "data" / "adapt.md"
 
         user_msg = (

core/framework/agents/queen/reference/anti_patterns.md

@@ -27,7 +27,9 @@
 ## GCU Errors
 15. **Manually wiring browser tools on event_loop nodes** — Use `node_type="gcu"` which auto-includes browser tools. Do NOT manually list browser tool names.
 16. **Using GCU nodes as regular graph nodes** — GCU nodes are subagents only. They must ONLY appear in `sub_agents=["gcu-node-id"]` and be invoked via `delegate_to_sub_agent()`. Never connect via edges or use as entry/terminal nodes.
+17. **Reusing the same GCU node ID for parallel tasks** — Each concurrent browser task needs a distinct GCU node ID (e.g. `gcu-site-a`, `gcu-site-b`). Two `delegate_to_sub_agent` calls with the same `agent_id` share a browser profile and will interfere with each other's pages.
+18. **Passing `profile=` in GCU tool calls** — Profile isolation for parallel subagents is automatic. The framework injects a unique profile per subagent via an asyncio `ContextVar`. Hardcoding `profile="default"` in a GCU system prompt breaks this isolation.
 
 ## Worker Agent Errors
-17. **Adding client-facing intake node to workers** — The queen owns intake. Workers should start with an autonomous processing node. Client-facing nodes in workers are for mid-execution review/approval only.
-18. **Putting `escalate` or `set_output` in NodeSpec `tools=[]`** — These are synthetic framework tools, auto-injected at runtime. Only list MCP tools from `list_agent_tools()`.
+19. **Adding client-facing intake node to workers** — The queen owns intake. Workers should start with an autonomous processing node. Client-facing nodes in workers are for mid-execution review/approval only.
+20. **Putting `escalate` or `set_output` in NodeSpec `tools=[]`** — These are synthetic framework tools, auto-injected at runtime. Only list MCP tools from `list_agent_tools()`.

core/framework/agents/queen/reference/file_templates.md

@@ -180,7 +180,7 @@ terminal_nodes = []  # Forever-alive
 # Module-level vars read by AgentRunner.load()
 conversation_mode = "continuous"
 identity_prompt = "You are a helpful agent."
-loop_config = {"max_iterations": 100, "max_tool_calls_per_turn": 20, "max_history_tokens": 32000}
+loop_config = {"max_iterations": 100, "max_tool_calls_per_turn": 20, "max_context_tokens": 32000}
 
 
 class MyAgent:
@@ -332,81 +332,46 @@ class MyAgent:
 default_agent = MyAgent()
 ```
 
-## agent.py — Async Entry Points Variant
+## triggers.json — Timer and Webhook Triggers
 
-When an agent needs timers, webhooks, or event-driven triggers, add
-`async_entry_points` and optionally `runtime_config` as module-level variables.
-These are IN ADDITION to the standard variables above.
+When an agent needs timers, webhooks, or event-driven triggers, create a
+`triggers.json` file in the agent's directory (alongside `agent.py`).
+The queen loads these at session start and the user can manage them via
+the `set_trigger` / `remove_trigger` tools at runtime.
 
-```python
-# Additional imports for async entry points
-from framework.graph.edge import GraphSpec, AsyncEntryPointSpec
-from framework.runtime.agent_runtime import (
-    AgentRuntime, AgentRuntimeConfig, create_agent_runtime,
-)
-
-# ... (goal, nodes, edges, entry_node, entry_points, etc. as above) ...
-
-# Async entry points — event-driven triggers
-async_entry_points = [
-    # Timer with cron: daily at 9am
-    AsyncEntryPointSpec(
-        id="daily-check",
-        name="Daily Check",
-        entry_node="process-node",
-        trigger_type="timer",
-        trigger_config={"cron": "0 9 * * *"},
-        isolation_level="shared",
-        max_concurrent=1,
-    ),
-    # Timer with fixed interval: every 20 minutes
-    AsyncEntryPointSpec(
-        id="scheduled-check",
-        name="Scheduled Check",
-        entry_node="process-node",
-        trigger_type="timer",
-        trigger_config={"interval_minutes": 20, "run_immediately": False},
-        isolation_level="shared",
-        max_concurrent=1,
-    ),
-    # Event: reacts to webhook events
-    AsyncEntryPointSpec(
-        id="webhook-event",
-        name="Webhook Event Handler",
-        entry_node="process-node",
-        trigger_type="event",
-        trigger_config={"event_types": ["webhook_received"]},
-        isolation_level="shared",
-        max_concurrent=10,
-    ),
+```json
+[
+  {
+    "id": "daily-check",
+    "name": "Daily Check",
+    "trigger_type": "timer",
+    "trigger_config": {"cron": "0 9 * * *"},
+    "task": "Run the daily check process"
+  },
+  {
+    "id": "scheduled-check",
+    "name": "Scheduled Check",
+    "trigger_type": "timer",
+    "trigger_config": {"interval_minutes": 20},
+    "task": "Run the scheduled check"
+  },
+  {
+    "id": "webhook-event",
+    "name": "Webhook Event Handler",
+    "trigger_type": "webhook",
+    "trigger_config": {"event_types": ["webhook_received"]},
+    "task": "Process incoming webhook event"
+  }
 ]
-
-# Webhook server config (only needed if using webhooks)
-runtime_config = AgentRuntimeConfig(
-    webhook_host="127.0.0.1",
-    webhook_port=8080,
-    webhook_routes=[
-        {
-            "source_id": "my-source",
-            "path": "/webhooks/my-source",
-            "methods": ["POST"],
-        },
-    ],
-)
 ```
 
-**Key rules for async entry points:**
-- `async_entry_points` is a list of `AsyncEntryPointSpec` (NOT `EntryPointSpec`)
-- `runtime_config` is `AgentRuntimeConfig` (NOT `RuntimeConfig` from config.py)
-- Valid trigger_types: `timer`, `event`, `webhook`, `manual`, `api`
-- Valid isolation_levels: `isolated`, `shared`, `synchronized`
+**Key rules for triggers.json:**
+- Valid trigger_types: `timer`, `webhook`
 - Timer trigger_config (cron): `{"cron": "0 9 * * *"}` — standard 5-field cron expression
-- Timer trigger_config (interval): `{"interval_minutes": float, "run_immediately": bool}`
-- Event trigger_config: `{"event_types": ["webhook_received"], "filter_stream": "...", "filter_node": "..."}`
-- Use `isolation_level="shared"` for async entry points that need to read
-  the primary session's memory (e.g., user-configured rules)
-- The `_build_graph()` method passes `async_entry_points` to GraphSpec
-- Reference: `exports/gmail_inbox_guardian/agent.py`
+- Timer trigger_config (interval): `{"interval_minutes": float}`
+- Each trigger must have a unique `id`
+- The `task` field describes what the worker should do when the trigger fires
+- Triggers are persisted back to `triggers.json` when modified via queen tools
 
 ## __init__.py
 
@@ -453,21 +418,6 @@ __all__ = [
 ]
 ```
 
-**If the agent uses async entry points**, also import and export:
-```python
-from .agent import (
-    ...,
-    async_entry_points,
-    runtime_config,  # Only if using webhooks
-)
-
-__all__ = [
-    ...,
-    "async_entry_points",
-    "runtime_config",
-]
-```
-
 ## __main__.py
 
 ```python

core/framework/agents/queen/reference/framework_guide.md

@@ -31,8 +31,7 @@ module-level variables via `getattr()`:
 | `conversation_mode` | no | not passed | Isolated mode (no context carryover) |
 | `identity_prompt` | no | not passed | No agent-level identity |
 | `loop_config` | no | `{}` | No iteration limits |
-| `async_entry_points` | no | `[]` | No async triggers (timers, webhooks, events) |
-| `runtime_config` | no | `None` | No webhook server |
+| `triggers.json` (file) | no | not present | No triggers (timers, webhooks) |
 
 **CRITICAL:** `__init__.py` MUST import and re-export ALL of these from
 `agent.py`. Missing exports silently fall back to defaults, causing
@@ -226,7 +225,7 @@ Only three valid keys:
 loop_config = {
     "max_iterations": 100,          # Max LLM turns per node visit
     "max_tool_calls_per_turn": 20,  # Max tool calls per LLM response
-    "max_history_tokens": 32000,    # Triggers conversation compaction
+    "max_context_tokens": 32000,    # Triggers conversation compaction
 }
 ```
 **INVALID keys** (do NOT use): `"strategy"`, `"mode"`, `"timeout"`,
@@ -257,44 +256,28 @@ Multiple ON_SUCCESS edges from same source → parallel execution via asyncio.ga
 
 Judge is the SOLE acceptance mechanism — no ad-hoc framework gating.
 
-## Async Entry Points (Webhooks, Timers, Events)
+## Triggers (Timers, Webhooks)
 
-For agents that react to external events, use `AsyncEntryPointSpec`:
+For agents that react to external events, create a `triggers.json` file
+in the agent's export directory:
 
-```python
-from framework.graph.edge import AsyncEntryPointSpec
-from framework.runtime.agent_runtime import AgentRuntimeConfig
-
-# Timer trigger (cron or interval)
-async_entry_points = [
-    AsyncEntryPointSpec(
-        id="daily-check",
-        name="Daily Check",
-        entry_node="process",
-        trigger_type="timer",
-        trigger_config={"cron": "0 9 * * *"},  # daily at 9am
-        isolation_level="shared",
-    )
+```json
+[
+  {
+    "id": "daily-check",
+    "name": "Daily Check",
+    "trigger_type": "timer",
+    "trigger_config": {"cron": "0 9 * * *"},
+    "task": "Run the daily check process"
+  }
 ]
-
-# Webhook server (optional)
-runtime_config = AgentRuntimeConfig(
-    webhook_host="127.0.0.1",
-    webhook_port=8080,
-    webhook_routes=[{"source_id": "gmail", "path": "/webhooks/gmail", "methods": ["POST"]}],
-)
 ```
 
 ### Key Fields
-- `trigger_type`: `"timer"`, `"event"`, `"webhook"`, `"manual"`
+- `trigger_type`: `"timer"` or `"webhook"`
 - `trigger_config`: `{"cron": "0 9 * * *"}` or `{"interval_minutes": 20}`
-- `isolation_level`: `"shared"` (recommended), `"isolated"`, `"synchronized"`
-- `event_types`: For event triggers, e.g., `["webhook_received"]`
-
-### Exports Required
-Both `async_entry_points` and `runtime_config` must be exported from `__init__.py`.
-
-See `exports/gmail_inbox_guardian/agent.py` for complete example.
+- `task`: describes what the worker should do when the trigger fires
+- Triggers can also be created/removed at runtime via `set_trigger` / `remove_trigger` queen tools
 
 ## Tool Discovery
 

core/framework/agents/queen/reference/gcu_guide.md

@@ -109,6 +109,45 @@ Key rules to bake into GCU node prompts:
 - Keep tool calls per turn ≤10
 - Tab isolation: when browser is already running, use `browser_open(background=true)` and pass `target_id` to every call
 
+## Multiple Concurrent GCU Subagents
+
+When a task can be parallelized across multiple sites or profiles, declare a distinct GCU
+node for each and invoke them all in the same LLM turn.  The framework batches all
+`delegate_to_sub_agent` calls made in one turn and runs them with `asyncio.gather`, so
+they execute concurrently — not sequentially.
+
+**Each GCU subagent automatically gets its own isolated browser context** — no `profile=`
+argument is needed in tool calls.  The framework derives a unique profile from the subagent's
+node ID and instance counter and injects it via an asyncio `ContextVar` before the subagent
+runs.
+
+### Example: three sites in parallel
+
+```python
+# Three distinct GCU nodes
+gcu_site_a = NodeSpec(id="gcu-site-a", node_type="gcu", ...)
+gcu_site_b = NodeSpec(id="gcu-site-b", node_type="gcu", ...)
+gcu_site_c = NodeSpec(id="gcu-site-c", node_type="gcu", ...)
+
+orchestrator = NodeSpec(
+    id="orchestrator",
+    node_type="event_loop",
+    sub_agents=["gcu-site-a", "gcu-site-b", "gcu-site-c"],
+    system_prompt="""\
+Call all three subagents in a single response to run them in parallel:
+  delegate_to_sub_agent(agent_id="gcu-site-a", task="Scrape prices from site A")
+  delegate_to_sub_agent(agent_id="gcu-site-b", task="Scrape prices from site B")
+  delegate_to_sub_agent(agent_id="gcu-site-c", task="Scrape prices from site C")
+""",
+)
+```
+
+**Rules:**
+- Use distinct node IDs for each concurrent task — sharing an ID shares the browser context.
+- The GCU node prompts do not need to mention `profile=`; isolation is automatic.
+- Cleanup is automatic at session end, but GCU nodes can call `browser_stop()` explicitly
+  if they want to release resources mid-run.
+
 ## GCU Anti-Patterns
 
 - Using `browser_screenshot` to read text (use `browser_snapshot`)

core/framework/agents/queen/ticket_receiver.py

@@ -1,8 +1,8 @@
 """Queen's ticket receiver entry point.
 
-When the Worker Health Judge emits a WORKER_ESCALATION_TICKET event on the
-shared EventBus, this entry point fires and routes to the ``ticket_triage``
-node, where the Queen deliberates and decides whether to notify the operator.
+When a WORKER_ESCALATION_TICKET event is emitted on the shared EventBus,
+this entry point fires and routes to the ``ticket_triage`` node, where the
+Queen deliberates and decides whether to notify the operator.
 
 Isolation level is ``isolated`` — the queen's triage memory is kept separate
 from the worker's shared memory. Each ticket triage runs in its own context.

core/framework/agents/worker_memory.py

@@ -0,0 +1,286 @@
+"""Worker per-run digest (run diary).
+
+Storage layout:
+    ~/.hive/agents/{agent_name}/runs/{run_id}/digest.md
+
+Each completed or failed worker run gets one digest file.  The queen reads
+these via get_worker_status(focus='diary') before digging into live runtime
+logs — the diary is a cheap, persistent record that survives across sessions.
+"""
+
+from __future__ import annotations
+
+import logging
+import traceback
+from collections import Counter
+from datetime import datetime
+from pathlib import Path
+from typing import TYPE_CHECKING, Any
+
+if TYPE_CHECKING:
+    from framework.runtime.event_bus import AgentEvent, EventBus
+
+logger = logging.getLogger(__name__)
+
+
+_DIGEST_SYSTEM = """\
+You maintain run digests for a worker agent.
+A run digest is a concise, factual record of a single task execution.
+
+Write 3-6 sentences covering:
+- What the worker was asked to do (the task/goal)
+- What approach it took and what tools it used
+- What the outcome was (success, partial, or failure — and why if relevant)
+- Any notable issues, retries, or escalations to the queen
+
+Write in third person past tense. Be direct and specific.
+Omit routine tool invocations unless the result matters.
+Output only the digest prose — no headings, no code fences.
+"""
+
+
+def _worker_runs_dir(agent_name: str) -> Path:
+    return Path.home() / ".hive" / "agents" / agent_name / "runs"
+
+
+def digest_path(agent_name: str, run_id: str) -> Path:
+    return _worker_runs_dir(agent_name) / run_id / "digest.md"
+
+
+def _collect_run_events(bus: EventBus, run_id: str, limit: int = 2000) -> list[AgentEvent]:
+    """Collect all events belonging to *run_id* from the bus history.
+
+    Strategy: find the EXECUTION_STARTED event that carries ``run_id``,
+    extract its ``execution_id``, then query the bus by that execution_id.
+    This works because TOOL_CALL_*, EDGE_TRAVERSED, NODE_STALLED etc. carry
+    execution_id but not run_id.
+
+    Falls back to a full-scan run_id filter when EXECUTION_STARTED is not
+    found (e.g. bus was rotated).
+    """
+    from framework.runtime.event_bus import EventType
+
+    # Pass 1: find execution_id via EXECUTION_STARTED with matching run_id
+    started = bus.get_history(event_type=EventType.EXECUTION_STARTED, limit=limit)
+    exec_id: str | None = None
+    for e in started:
+        if getattr(e, "run_id", None) == run_id and e.execution_id:
+            exec_id = e.execution_id
+            break
+
+    if exec_id:
+        return bus.get_history(execution_id=exec_id, limit=limit)
+
+    # Fallback: scan all events and match by run_id attribute
+    return [e for e in bus.get_history(limit=limit) if getattr(e, "run_id", None) == run_id]
+
+
+def _build_run_context(
+    events: list[AgentEvent],
+    outcome_event: AgentEvent | None,
+) -> str:
+    """Assemble a plain-text run context string for the digest LLM call."""
+    from framework.runtime.event_bus import EventType
+
+    # Reverse so events are in chronological order
+    events_chron = list(reversed(events))
+
+    lines: list[str] = []
+
+    # Task input from EXECUTION_STARTED
+    started = [e for e in events_chron if e.type == EventType.EXECUTION_STARTED]
+    if started:
+        inp = started[0].data.get("input", {})
+        if inp:
+            lines.append(f"Task input: {str(inp)[:400]}")
+
+    # Duration (elapsed so far if no outcome yet)
+    ref_ts = outcome_event.timestamp if outcome_event else datetime.utcnow()
+    if started:
+        elapsed = (ref_ts - started[0].timestamp).total_seconds()
+        m, s = divmod(int(elapsed), 60)
+        lines.append(f"Duration so far: {m}m {s}s" if m else f"Duration so far: {s}s")
+
+    # Outcome
+    if outcome_event is None:
+        lines.append("Status: still running (mid-run snapshot)")
+    elif outcome_event.type == EventType.EXECUTION_COMPLETED:
+        out = outcome_event.data.get("output", {})
+        out_str = f"Outcome: completed. Output: {str(out)[:300]}"
+        lines.append(out_str if out else "Outcome: completed.")
+    else:
+        err = outcome_event.data.get("error", "")
+        lines.append(f"Outcome: failed. Error: {str(err)[:300]}" if err else "Outcome: failed.")
+
+    # Node path (edge traversals)
+    edges = [e for e in events_chron if e.type == EventType.EDGE_TRAVERSED]
+    if edges:
+        parts = [
+            f"{e.data.get('source_node', '?')}->{e.data.get('target_node', '?')}"
+            for e in edges[-20:]
+        ]
+        lines.append(f"Node path: {', '.join(parts)}")
+
+    # Tools used
+    tool_events = [e for e in events_chron if e.type == EventType.TOOL_CALL_COMPLETED]
+    if tool_events:
+        names = [e.data.get("tool_name", "?") for e in tool_events]
+        counts = Counter(names)
+        summary = ", ".join(f"{name}×{n}" if n > 1 else name for name, n in counts.most_common())
+        lines.append(f"Tools used: {summary}")
+        # Note any tool errors
+        errors = [e for e in tool_events if e.data.get("is_error")]
+        if errors:
+            err_names = Counter(e.data.get("tool_name", "?") for e in errors)
+            lines.append(f"Tool errors: {dict(err_names)}")
+
+    # Issues
+    issue_map = {
+        EventType.NODE_STALLED: "stall",
+        EventType.NODE_TOOL_DOOM_LOOP: "doom loop",
+        EventType.CONSTRAINT_VIOLATION: "constraint violation",
+        EventType.NODE_RETRY: "retry",
+    }
+    issue_parts: list[str] = []
+    for evt_type, label in issue_map.items():
+        n = sum(1 for e in events_chron if e.type == evt_type)
+        if n:
+            issue_parts.append(f"{n} {label}(s)")
+    if issue_parts:
+        lines.append(f"Issues: {', '.join(issue_parts)}")
+
+    # Escalations to queen
+    escalations = [e for e in events_chron if e.type == EventType.ESCALATION_REQUESTED]
+    if escalations:
+        lines.append(f"Escalations to queen: {len(escalations)}")
+
+    # Final LLM output snippet (last LLM_TEXT_DELTA snapshot)
+    text_events = [e for e in reversed(events_chron) if e.type == EventType.LLM_TEXT_DELTA]
+    if text_events:
+        snapshot = text_events[0].data.get("snapshot", "") or ""
+        if snapshot:
+            lines.append(f"Final LLM output: {snapshot[-400:].strip()}")
+
+    return "\n".join(lines)
+
+
+async def consolidate_worker_run(
+    agent_name: str,
+    run_id: str,
+    outcome_event: AgentEvent | None,
+    bus: EventBus,
+    llm: Any,
+) -> None:
+    """Write (or overwrite) the digest for a worker run.
+
+    Called fire-and-forget either:
+    - After EXECUTION_COMPLETED / EXECUTION_FAILED (outcome_event set, final write)
+    - Periodically during a run on a cooldown timer (outcome_event=None, mid-run snapshot)
+
+    The digest file is always overwritten so each call produces the freshest view.
+    The final completion/failure call supersedes any mid-run snapshot.
+
+    Args:
+        agent_name:    Worker agent directory name (determines storage path).
+        run_id:        The run ID.
+        outcome_event: EXECUTION_COMPLETED or EXECUTION_FAILED event, or None for
+                       a mid-run snapshot.
+        bus:           The session EventBus (shared queen + worker).
+        llm:           LLMProvider with an acomplete() method.
+    """
+    try:
+        events = _collect_run_events(bus, run_id)
+        run_context = _build_run_context(events, outcome_event)
+        if not run_context:
+            logger.debug("worker_memory: no events for run %s, skipping digest", run_id)
+            return
+
+        is_final = outcome_event is not None
+        logger.info(
+            "worker_memory: generating %s digest for run %s ...",
+            "final" if is_final else "mid-run",
+            run_id,
+        )
+
+        from framework.agents.queen.config import default_config
+
+        resp = await llm.acomplete(
+            messages=[{"role": "user", "content": run_context}],
+            system=_DIGEST_SYSTEM,
+            max_tokens=min(default_config.max_tokens, 512),
+        )
+        digest_text = (resp.content or "").strip()
+        if not digest_text:
+            logger.warning("worker_memory: LLM returned empty digest for run %s", run_id)
+            return
+
+        path = digest_path(agent_name, run_id)
+        path.parent.mkdir(parents=True, exist_ok=True)
+
+        from framework.runtime.event_bus import EventType
+
+        ts = (outcome_event.timestamp if outcome_event else datetime.utcnow()).strftime(
+            "%Y-%m-%d %H:%M"
+        )
+        if outcome_event is None:
+            status = "running"
+        elif outcome_event.type == EventType.EXECUTION_COMPLETED:
+            status = "completed"
+        else:
+            status = "failed"
+
+        path.write_text(
+            f"# {run_id}\n\n**{ts}** | {status}\n\n{digest_text}\n",
+            encoding="utf-8",
+        )
+        logger.info(
+            "worker_memory: %s digest written for run %s (%d chars)",
+            status,
+            run_id,
+            len(digest_text),
+        )
+
+    except Exception:
+        tb = traceback.format_exc()
+        logger.exception("worker_memory: digest failed for run %s", run_id)
+        # Persist the error so it's findable without log access
+        error_path = _worker_runs_dir(agent_name) / run_id / "digest_error.txt"
+        try:
+            error_path.parent.mkdir(parents=True, exist_ok=True)
+            error_path.write_text(
+                f"run_id: {run_id}\ntime: {datetime.now().isoformat()}\n\n{tb}",
+                encoding="utf-8",
+            )
+        except Exception:
+            pass
+
+
+def read_recent_digests(agent_name: str, max_runs: int = 5) -> list[tuple[str, str]]:
+    """Return recent run digests as [(run_id, content), ...], newest first.
+
+    Args:
+        agent_name: Worker agent directory name.
+        max_runs:   Maximum number of digests to return.
+
+    Returns:
+        List of (run_id, digest_content) tuples, ordered newest first.
+    """
+    runs_dir = _worker_runs_dir(agent_name)
+    if not runs_dir.exists():
+        return []
+
+    digest_files = sorted(
+        runs_dir.glob("*/digest.md"),
+        key=lambda p: p.stat().st_mtime,
+        reverse=True,
+    )[:max_runs]
+
+    result: list[tuple[str, str]] = []
+    for f in digest_files:
+        try:
+            content = f.read_text(encoding="utf-8").strip()
+            if content:
+                result.append((f.parent.name, content))
+        except OSError:
+            continue
+    return result

core/framework/config.py

@@ -19,6 +19,10 @@ from framework.graph.edge import DEFAULT_MAX_TOKENS
 # ---------------------------------------------------------------------------
 
 HIVE_CONFIG_FILE = Path.home() / ".hive" / "configuration.json"
+
+# Hive LLM router endpoint (Anthropic-compatible).
+# litellm's Anthropic handler appends /v1/messages, so this is just the base host.
+HIVE_LLM_ENDPOINT = "https://api.adenhq.com"
 logger = logging.getLogger(__name__)
 
 
@@ -47,7 +51,13 @@ def get_preferred_model() -> str:
     """Return the user's preferred LLM model string (e.g. 'anthropic/claude-sonnet-4-20250514')."""
     llm = get_hive_config().get("llm", {})
     if llm.get("provider") and llm.get("model"):
-        return f"{llm['provider']}/{llm['model']}"
+        provider = str(llm["provider"])
+        model = str(llm["model"]).strip()
+        # OpenRouter quickstart stores raw model IDs; tolerate pasted "openrouter/<id>" too.
+        if provider.lower() == "openrouter" and model.lower().startswith("openrouter/"):
+            model = model[len("openrouter/") :]
+        if model:
+            return f"{provider}/{model}"
     return "anthropic/claude-sonnet-4-20250514"
 
 
@@ -56,6 +66,15 @@ def get_max_tokens() -> int:
     return get_hive_config().get("llm", {}).get("max_tokens", DEFAULT_MAX_TOKENS)
 
 
+DEFAULT_MAX_CONTEXT_TOKENS = 32_000
+OPENROUTER_API_BASE = "https://openrouter.ai/api/v1"
+
+
+def get_max_context_tokens() -> int:
+    """Return the configured max_context_tokens, falling back to DEFAULT_MAX_CONTEXT_TOKENS."""
+    return get_hive_config().get("llm", {}).get("max_context_tokens", DEFAULT_MAX_CONTEXT_TOKENS)
+
+
 def get_api_key() -> str | None:
     """Return the API key, supporting env var, Claude Code subscription, Codex, and ZAI Code.
 
@@ -113,6 +132,14 @@ def get_gcu_enabled() -> bool:
     return get_hive_config().get("gcu_enabled", True)
 
 
+def get_gcu_viewport_scale() -> float:
+    """Return GCU viewport scale factor (0.1-1.0), default 0.8."""
+    scale = get_hive_config().get("gcu_viewport_scale", 0.8)
+    if isinstance(scale, (int, float)) and 0.1 <= scale <= 1.0:
+        return float(scale)
+    return 0.8
+
+
 def get_api_base() -> str | None:
     """Return the api_base URL for OpenAI-compatible endpoints, if configured."""
     llm = get_hive_config().get("llm", {})
@@ -122,7 +149,11 @@ def get_api_base() -> str | None:
     if llm.get("use_kimi_code_subscription"):
         # Kimi Code uses an Anthropic-compatible endpoint (no /v1 suffix).
         return "https://api.kimi.com/coding"
-    return llm.get("api_base")
+    if llm.get("api_base"):
+        return llm["api_base"]
+    if str(llm.get("provider", "")).lower() == "openrouter":
+        return OPENROUTER_API_BASE
+    return None
 
 
 def get_llm_extra_kwargs() -> dict[str, Any]:
@@ -178,6 +209,7 @@ class RuntimeConfig:
     model: str = field(default_factory=get_preferred_model)
     temperature: float = 0.7
     max_tokens: int = field(default_factory=get_max_tokens)
+    max_context_tokens: int = field(default_factory=get_max_context_tokens)
     api_key: str | None = field(default_factory=get_api_key)
     api_base: str | None = field(default_factory=get_api_base)
     extra_kwargs: dict[str, Any] = field(default_factory=get_llm_extra_kwargs)

core/framework/credentials/key_storage.py

@@ -142,17 +142,27 @@ def save_aden_api_key(key: str) -> None:
     os.environ[ADEN_ENV_VAR] = key
 
 
-def delete_aden_api_key() -> None:
-    """Remove ADEN_API_KEY from the encrypted store and ``os.environ``."""
+def delete_aden_api_key() -> bool:
+    """Remove ADEN_API_KEY from the encrypted store and ``os.environ``.
+
+    Returns True if the key existed and was deleted, False otherwise.
+    """
+    deleted = False
     try:
         from .storage import EncryptedFileStorage
 
         storage = EncryptedFileStorage()
-        storage.delete(ADEN_CREDENTIAL_ID)
+        deleted = storage.delete(ADEN_CREDENTIAL_ID)
+    except (FileNotFoundError, PermissionError) as e:
+        logger.debug("Could not delete %s from encrypted store: %s", ADEN_CREDENTIAL_ID, e)
     except Exception:
-        logger.debug("Could not delete %s from encrypted store", ADEN_CREDENTIAL_ID)
-
+        logger.warning(
+            "Unexpected error deleting %s from encrypted store",
+            ADEN_CREDENTIAL_ID,
+            exc_info=True,
+        )
     os.environ.pop(ADEN_ENV_VAR, None)
+    return deleted
 
 
 # ---------------------------------------------------------------------------
@@ -167,8 +177,10 @@ def _read_credential_key_file() -> str | None:
             value = CREDENTIAL_KEY_PATH.read_text(encoding="utf-8").strip()
             if value:
                 return value
+    except (FileNotFoundError, PermissionError) as e:
+        logger.debug("Could not read %s: %s", CREDENTIAL_KEY_PATH, e)
     except Exception:
-        logger.debug("Could not read %s", CREDENTIAL_KEY_PATH)
+        logger.warning("Unexpected error reading %s", CREDENTIAL_KEY_PATH, exc_info=True)
     return None
 
 
@@ -196,6 +208,12 @@ def _read_aden_from_encrypted_store() -> str | None:
         cred = storage.load(ADEN_CREDENTIAL_ID)
         if cred:
             return cred.get_key("api_key")
+    except (FileNotFoundError, PermissionError, KeyError) as e:
+        logger.debug("Could not load %s from encrypted store: %s", ADEN_CREDENTIAL_ID, e)
     except Exception:
-        logger.debug("Could not load %s from encrypted store", ADEN_CREDENTIAL_ID)
+        logger.warning(
+            "Unexpected error loading %s from encrypted store",
+            ADEN_CREDENTIAL_ID,
+            exc_info=True,
+        )
     return None

core/framework/credentials/validation.py

@@ -51,6 +51,16 @@ def ensure_credential_key_env() -> None:
                     if found and value:
                         os.environ[var_name] = value
                         logger.debug("Loaded %s from shell config", var_name)
+        # Also load the currently configured LLM env var even if it's not in CREDENTIAL_SPECS.
+        # This keeps quickstart-written keys available to fresh processes on Unix shells.
+        from framework.config import get_hive_config
+
+        llm_env_var = str(get_hive_config().get("llm", {}).get("api_key_env_var", "")).strip()
+        if llm_env_var and not os.environ.get(llm_env_var):
+            found, value = check_env_var_in_shell_config(llm_env_var)
+            if found and value:
+                os.environ[llm_env_var] = value
+                logger.debug("Loaded configured LLM env var %s from shell config", llm_env_var)
     except ImportError:
         pass
 

core/framework/graph/conversation.py

@@ -307,13 +307,13 @@ class NodeConversation:
     def __init__(
         self,
         system_prompt: str = "",
-        max_history_tokens: int = 32000,
+        max_context_tokens: int = 32000,
         compaction_threshold: float = 0.8,
         output_keys: list[str] | None = None,
         store: ConversationStore | None = None,
     ) -> None:
         self._system_prompt = system_prompt
-        self._max_history_tokens = max_history_tokens
+        self._max_context_tokens = max_context_tokens
         self._compaction_threshold = compaction_threshold
         self._output_keys = output_keys
         self._store = store
@@ -525,16 +525,16 @@ class NodeConversation:
         self._last_api_input_tokens = actual_input_tokens
 
     def usage_ratio(self) -> float:
-        """Current token usage as a fraction of *max_history_tokens*.
+        """Current token usage as a fraction of *max_context_tokens*.
 
-        Returns 0.0 when ``max_history_tokens`` is zero (unlimited).
+        Returns 0.0 when ``max_context_tokens`` is zero (unlimited).
         """
-        if self._max_history_tokens <= 0:
+        if self._max_context_tokens <= 0:
             return 0.0
-        return self.estimate_tokens() / self._max_history_tokens
+        return self.estimate_tokens() / self._max_context_tokens
 
     def needs_compaction(self) -> bool:
-        return self.estimate_tokens() >= self._max_history_tokens * self._compaction_threshold
+        return self.estimate_tokens() >= self._max_context_tokens * self._compaction_threshold
 
     # --- Output-key extraction ---------------------------------------------
 
@@ -612,6 +612,11 @@ class NodeConversation:
                 continue  # never prune errors
             if msg.content.startswith("[Pruned tool result"):
                 continue  # already pruned
+            # Tiny results (set_output acks, confirmations) — pruning
+            # saves negligible space but makes the LLM think the call
+            # failed, causing costly retries.
+            if len(msg.content) < 100:
+                continue
 
             # Phase-aware: protect current phase messages
             if self._current_phase and msg.phase_id == self._current_phase:
@@ -901,8 +906,7 @@ class NodeConversation:
             full_path = str((spill_path / conv_filename).resolve())
             ref_parts.append(
                 f"[Previous conversation saved to '{full_path}'. "
-                f"Use load_data('{conv_filename}'), read_file('{full_path}'), "
-                f"or run_command('cat \"{full_path}\"') to review if needed.]"
+                f"Use load_data('{conv_filename}') to review if needed.]"
             )
         elif not collapsed_msgs:
             ref_parts.append("[Previous freeform messages compacted.]")
@@ -1029,7 +1033,7 @@ class NodeConversation:
         await self._store.write_meta(
             {
                 "system_prompt": self._system_prompt,
-                "max_history_tokens": self._max_history_tokens,
+                "max_context_tokens": self._max_context_tokens,
                 "compaction_threshold": self._compaction_threshold,
                 "output_keys": self._output_keys,
             }
@@ -1062,7 +1066,7 @@ class NodeConversation:
 
         conv = cls(
             system_prompt=meta.get("system_prompt", ""),
-            max_history_tokens=meta.get("max_history_tokens", 32000),
+            max_context_tokens=meta.get("max_context_tokens", 32000),
             compaction_threshold=meta.get("compaction_threshold", 0.8),
             output_keys=meta.get("output_keys"),
             store=store,

core/framework/graph/conversation_judge.py

@@ -37,7 +37,7 @@ async def evaluate_phase_completion(
     phase_description: str,
     success_criteria: str,
     accumulator_state: dict[str, Any],
-    max_history_tokens: int = 8_196,
+    max_context_tokens: int = 8_196,
 ) -> PhaseVerdict:
     """Level 2 judge: read the conversation and evaluate quality.
 
@@ -50,7 +50,7 @@ async def evaluate_phase_completion(
         phase_description: Description of the phase
         success_criteria: Natural-language criteria for phase completion
         accumulator_state: Current output key values
-        max_history_tokens: Main conversation token budget (judge gets 20%)
+        max_context_tokens: Main conversation token budget (judge gets 20%)
 
     Returns:
         PhaseVerdict with action and optional feedback
@@ -89,7 +89,7 @@ FEEDBACK: (reason if RETRY, empty if ACCEPT)"""
         response = await llm.acomplete(
             messages=[{"role": "user", "content": user_prompt}],
             system=system_prompt,
-            max_tokens=max(1024, max_history_tokens // 5),
+            max_tokens=max(1024, max_context_tokens // 5),
             max_retries=1,
         )
         if not response.content or not response.content.strip():

core/framework/graph/edge.py

@@ -322,7 +322,11 @@ class AsyncEntryPointSpec(BaseModel):
 
     id: str = Field(description="Unique identifier for this entry point")
     name: str = Field(description="Human-readable name")
-    entry_node: str = Field(description="Node ID to start execution from")
+    entry_node: str = Field(
+        default="",
+        description="Deprecated: Node ID to start execution from. "
+        "Triggers are graph-level; worker always enters at GraphSpec.entry_node.",
+    )
     trigger_type: str = Field(
         default="manual",
         description="How this entry point is triggered: webhook, api, timer, event, manual",
@@ -331,6 +335,10 @@ class AsyncEntryPointSpec(BaseModel):
         default_factory=dict,
         description="Trigger-specific configuration (e.g., webhook URL, timer interval)",
     )
+    task: str = Field(
+        default="",
+        description="Worker task string when this trigger fires autonomously",
+    )
     isolation_level: str = Field(
         default="shared", description="State isolation: isolated, shared, or synchronized"
     )
@@ -368,28 +376,8 @@ class GraphSpec(BaseModel):
             edges=[...],
         )
 
-    For multi-entry-point agents (concurrent streams):
-        GraphSpec(
-            id="support-agent-graph",
-            goal_id="support-001",
-            entry_node="process-webhook",  # Default entry
-            async_entry_points=[
-                AsyncEntryPointSpec(
-                    id="webhook",
-                    name="Zendesk Webhook",
-                    entry_node="process-webhook",
-                    trigger_type="webhook",
-                ),
-                AsyncEntryPointSpec(
-                    id="api",
-                    name="API Handler",
-                    entry_node="process-request",
-                    trigger_type="api",
-                ),
-            ],
-            nodes=[...],
-            edges=[...],
-        )
+    Triggers (timer, webhook, event) are now defined in ``triggers.json``
+    alongside the agent directory, not embedded in the graph spec.
     """
 
     id: str
@@ -402,12 +390,6 @@ class GraphSpec(BaseModel):
         default_factory=dict,
         description="Named entry points for resuming execution. Format: {name: node_id}",
     )
-    async_entry_points: list[AsyncEntryPointSpec] = Field(
-        default_factory=list,
-        description=(
-            "Asynchronous entry points for concurrent execution streams (used with AgentRuntime)"
-        ),
-    )
     terminal_nodes: list[str] = Field(
         default_factory=list, description="IDs of nodes that end execution"
     )
@@ -486,17 +468,6 @@ class GraphSpec(BaseModel):
                 return node
         return None
 
-    def has_async_entry_points(self) -> bool:
-        """Check if this graph uses async entry points (multi-stream execution)."""
-        return len(self.async_entry_points) > 0
-
-    def get_async_entry_point(self, entry_point_id: str) -> AsyncEntryPointSpec | None:
-        """Get an async entry point by ID."""
-        for ep in self.async_entry_points:
-            if ep.id == entry_point_id:
-                return ep
-        return None
-
     def get_outgoing_edges(self, node_id: str) -> list[EdgeSpec]:
         """Get all edges leaving a node, sorted by priority."""
         edges = [e for e in self.edges if e.source == node_id]
@@ -587,37 +558,6 @@ class GraphSpec(BaseModel):
         if not self.get_node(self.entry_node):
             errors.append(f"Entry node '{self.entry_node}' not found")
 
-        # Check async entry points
-        seen_entry_ids = set()
-        for entry_point in self.async_entry_points:
-            # Check for duplicate IDs
-            if entry_point.id in seen_entry_ids:
-                errors.append(f"Duplicate async entry point ID: '{entry_point.id}'")
-            seen_entry_ids.add(entry_point.id)
-
-            # Check entry node exists
-            if not self.get_node(entry_point.entry_node):
-                errors.append(
-                    f"Async entry point '{entry_point.id}' references "
-                    f"missing node '{entry_point.entry_node}'"
-                )
-
-            # Validate isolation level
-            valid_isolation = {"isolated", "shared", "synchronized"}
-            if entry_point.isolation_level not in valid_isolation:
-                errors.append(
-                    f"Async entry point '{entry_point.id}' has invalid isolation_level "
-                    f"'{entry_point.isolation_level}'. Valid: {valid_isolation}"
-                )
-
-            # Validate trigger type
-            valid_triggers = {"webhook", "api", "timer", "event", "manual"}
-            if entry_point.trigger_type not in valid_triggers:
-                errors.append(
-                    f"Async entry point '{entry_point.id}' has invalid trigger_type "
-                    f"'{entry_point.trigger_type}'. Valid: {valid_triggers}"
-                )
-
         # Check terminal nodes exist
         for term in self.terminal_nodes:
             if not self.get_node(term):
@@ -646,10 +586,6 @@ class GraphSpec(BaseModel):
         for entry_point_node in self.entry_points.values():
             to_visit.append(entry_point_node)
 
-        # Add all async entry points as valid starting points
-        for async_entry in self.async_entry_points:
-            to_visit.append(async_entry.entry_node)
-
         # Traverse from all entry points
         while to_visit:
             current = to_visit.pop()
@@ -666,18 +602,10 @@ class GraphSpec(BaseModel):
                 for sub_agent_id in sub_agents:
                     reachable.add(sub_agent_id)
 
-        # Build set of async entry point nodes for quick lookup
-        async_entry_nodes = {ep.entry_node for ep in self.async_entry_points}
-
         for node in self.nodes:
             if node.id not in reachable:
-                # Skip if node is a pause node, entry point target, or async entry
-                # (pause/resume architecture and async entry points make reachable)
-                if (
-                    node.id in self.pause_nodes
-                    or node.id in self.entry_points.values()
-                    or node.id in async_entry_nodes
-                ):
+                # Skip if node is a pause node or entry point target
+                if node.id in self.pause_nodes or node.id in self.entry_points.values():
                     continue
                 errors.append(f"Node '{node.id}' is unreachable from entry")
 

core/framework/graph/executor.py

@@ -27,11 +27,24 @@ from framework.graph.node import (
     SharedMemory,
 )
 from framework.graph.validator import OutputValidator
-from framework.llm.provider import LLMProvider, Tool
+from framework.llm.provider import LLMProvider, Tool, ToolUse
 from framework.observability import set_trace_context
 from framework.runtime.core import Runtime
 from framework.schemas.checkpoint import Checkpoint
 from framework.storage.checkpoint_store import CheckpointStore
+from framework.utils.io import atomic_write
+
+logger = logging.getLogger(__name__)
+
+
+def _default_max_context_tokens() -> int:
+    """Resolve max_context_tokens from global config, falling back to 32000."""
+    try:
+        from framework.config import get_max_context_tokens
+
+        return get_max_context_tokens()
+    except Exception:
+        return 32_000
 
 
 @dataclass
@@ -138,6 +151,9 @@ class GraphExecutor:
         tool_provider_map: dict[str, str] | None = None,
         dynamic_tools_provider: Callable | None = None,
         dynamic_prompt_provider: Callable | None = None,
+        iteration_metadata_provider: Callable | None = None,
+        skills_catalog_prompt: str = "",
+        protocols_prompt: str = "",
     ):
         """
         Initialize the executor.
@@ -163,6 +179,8 @@ class GraphExecutor:
                 tool list (for mode switching)
             dynamic_prompt_provider: Optional callback returning current
                 system prompt (for phase switching)
+            skills_catalog_prompt: Available skills catalog for system prompt
+            protocols_prompt: Default skill operational protocols for system prompt
         """
         self.runtime = runtime
         self.llm = llm
@@ -183,6 +201,21 @@ class GraphExecutor:
         self.tool_provider_map = tool_provider_map
         self.dynamic_tools_provider = dynamic_tools_provider
         self.dynamic_prompt_provider = dynamic_prompt_provider
+        self.iteration_metadata_provider = iteration_metadata_provider
+        self.skills_catalog_prompt = skills_catalog_prompt
+        self.protocols_prompt = protocols_prompt
+
+        if protocols_prompt:
+            self.logger.info(
+                "GraphExecutor[%s] received protocols_prompt (%d chars)",
+                stream_id,
+                len(protocols_prompt),
+            )
+        else:
+            self.logger.warning(
+                "GraphExecutor[%s] received EMPTY protocols_prompt",
+                stream_id,
+            )
 
         # Parallel execution settings
         self.enable_parallel_execution = enable_parallel_execution
@@ -212,11 +245,11 @@ class GraphExecutor:
         """
         if not self._storage_path:
             return
+        state_path = self._storage_path / "state.json"
         try:
             import json as _json
             from datetime import datetime
 
-            state_path = self._storage_path / "state.json"
             if state_path.exists():
                 state_data = _json.loads(state_path.read_text(encoding="utf-8"))
             else:
@@ -239,9 +272,14 @@ class GraphExecutor:
             state_data["memory"] = memory_snapshot
             state_data["memory_keys"] = list(memory_snapshot.keys())
 
-            state_path.write_text(_json.dumps(state_data, indent=2), encoding="utf-8")
+            with atomic_write(state_path, encoding="utf-8") as f:
+                _json.dump(state_data, f, indent=2)
         except Exception:
-            pass  # Best-effort — never block execution
+            logger.warning(
+                "Failed to persist progress state to %s",
+                state_path,
+                exc_info=True,
+            )
 
     def _validate_tools(self, graph: GraphSpec) -> list[str]:
         """
@@ -330,7 +368,7 @@ class GraphExecutor:
                 _depth,
             )
         else:
-            max_tokens = getattr(conversation, "_max_history_tokens", 32000)
+            max_tokens = getattr(conversation, "_max_context_tokens", 32000)
             target_tokens = max_tokens // 2
             target_chars = target_tokens * 4
 
@@ -403,6 +441,14 @@ class GraphExecutor:
         )
         return s1 + "\n\n" + s2
 
+    def _get_runtime_log_session_id(self) -> str:
+        """Return the session-backed execution ID for runtime logging, if any."""
+        if not self._storage_path:
+            return ""
+        if self._storage_path.parent.name != "sessions":
+            return ""
+        return self._storage_path.name
+
     async def execute(
         self,
         graph: GraphSpec,
@@ -696,10 +742,7 @@ class GraphExecutor:
         )
 
         if self.runtime_logger:
-            # Extract session_id from storage_path if available (for unified sessions)
-            session_id = ""
-            if self._storage_path and self._storage_path.name.startswith("session_"):
-                session_id = self._storage_path.name
+            session_id = self._get_runtime_log_session_id()
             self.runtime_logger.start_run(goal_id=goal.id, session_id=session_id)
 
         self.logger.info(f"🚀 Starting execution: {goal.name}")
@@ -925,6 +968,33 @@ class GraphExecutor:
                 self.logger.info("   Executing...")
                 result = await node_impl.execute(ctx)
 
+                # GCU tab cleanup: stop the browser profile after a top-level GCU node
+                # finishes so tabs don't accumulate. Mirrors the subagent cleanup in
+                # EventLoopNode._execute_subagent().
+                if node_spec.node_type == "gcu" and self.tool_executor is not None:
+                    try:
+                        from gcu.browser.session import (
+                            _active_profile as _gcu_profile_var,
+                        )
+
+                        _gcu_profile = _gcu_profile_var.get()
+                        _stop_use = ToolUse(
+                            id="gcu-cleanup",
+                            name="browser_stop",
+                            input={"profile": _gcu_profile},
+                        )
+                        _stop_result = self.tool_executor(_stop_use)
+                        if asyncio.iscoroutine(_stop_result) or asyncio.isfuture(_stop_result):
+                            await _stop_result
+                    except ImportError:
+                        pass  # GCU not installed
+                    except Exception as _gcu_exc:
+                        logger.warning(
+                            "GCU browser_stop failed for profile %r: %s",
+                            _gcu_profile,
+                            _gcu_exc,
+                        )
+
                 # Emit node-completed event (skip event_loop nodes)
                 if self._event_bus and node_spec.node_type != "event_loop":
                     await self._event_bus.emit_node_loop_completed(
@@ -1350,6 +1420,7 @@ class GraphExecutor:
                     next_spec = graph.get_node(current_node_id)
                     if next_spec and next_spec.node_type == "event_loop":
                         from framework.graph.prompt_composer import (
+                            EXECUTION_SCOPE_PREAMBLE,
                             build_accounts_prompt,
                             build_narrative,
                             build_transition_marker,
@@ -1389,9 +1460,14 @@ class GraphExecutor:
                             )
 
                         # Compose new system prompt (Layer 1 + 2 + 3 + accounts)
+                        # Prepend scope preamble to focus so the LLM stays
+                        # within this node's responsibility.
+                        _focus = next_spec.system_prompt
+                        if next_spec.output_keys and _focus:
+                            _focus = f"{EXECUTION_SCOPE_PREAMBLE}\n\n{_focus}"
                         new_system = compose_system_prompt(
                             identity_prompt=getattr(graph, "identity_prompt", None),
-                            focus_prompt=next_spec.system_prompt,
+                            focus_prompt=_focus,
                             narrative=narrative,
                             accounts_prompt=_node_accounts,
                         )
@@ -1753,10 +1829,31 @@ class GraphExecutor:
             if node_spec.tools:
                 available_tools = [t for t in self.tools if t.name in node_spec.tools]
 
-        # Create scoped memory view
+        # Create scoped memory view.
+        # When permissions are restricted (non-empty key lists), auto-include
+        # _-prefixed keys used by default skill protocols so agents can read/write
+        # operational state (e.g. _working_notes, _batch_ledger) regardless of
+        # what the node declares.  When key lists are empty (unrestricted), leave
+        # unchanged — empty means "allow all".
+        read_keys = list(node_spec.input_keys)
+        write_keys = list(node_spec.output_keys)
+        # Only extend lists that were already restricted (non-empty).
+        # Empty means "allow all" — adding keys would accidentally
+        # activate the permission check and block legitimate reads/writes.
+        if read_keys or write_keys:
+            from framework.skills.defaults import SHARED_MEMORY_KEYS as _skill_keys
+
+            existing_underscore = [k for k in memory._data if k.startswith("_")]
+            extra_keys = set(_skill_keys) | set(existing_underscore)
+            for k in extra_keys:
+                if read_keys and k not in read_keys:
+                    read_keys.append(k)
+                if write_keys and k not in write_keys:
+                    write_keys.append(k)
+
         scoped_memory = memory.with_permissions(
-            read_keys=node_spec.input_keys,
-            write_keys=node_spec.output_keys,
+            read_keys=read_keys,
+            write_keys=write_keys,
         )
 
         # Build per-node accounts prompt (filtered to this node's tools)
@@ -1799,6 +1896,9 @@ class GraphExecutor:
             shared_node_registry=self.node_registry,  # For subagent escalation routing
             dynamic_tools_provider=self.dynamic_tools_provider,
             dynamic_prompt_provider=self.dynamic_prompt_provider,
+            iteration_metadata_provider=self.iteration_metadata_provider,
+            skills_catalog_prompt=self.skills_catalog_prompt,
+            protocols_prompt=self.protocols_prompt,
         )
 
     VALID_NODE_TYPES = {
@@ -1872,7 +1972,7 @@ class GraphExecutor:
                     max_tool_calls_per_turn=lc.get("max_tool_calls_per_turn", 30),
                     tool_call_overflow_margin=lc.get("tool_call_overflow_margin", 0.5),
                     stall_detection_threshold=lc.get("stall_detection_threshold", 3),
-                    max_history_tokens=lc.get("max_history_tokens", 32000),
+                    max_context_tokens=lc.get("max_context_tokens", _default_max_context_tokens()),
                     max_tool_result_chars=lc.get("max_tool_result_chars", 30_000),
                     spillover_dir=spillover,
                     hooks=lc.get("hooks", {}),
@@ -2039,6 +2139,10 @@ class GraphExecutor:
                 edge=edge,
             )
 
+        # Track which branch wrote which key for memory conflict detection
+        fanout_written_keys: dict[str, str] = {}  # key -> branch_id that wrote it
+        fanout_keys_lock = asyncio.Lock()
+
         self.logger.info(f"   ⑂ Fan-out: executing {len(branches)} branches in parallel")
         for branch in branches.values():
             target_spec = graph.get_node(branch.node_id)
@@ -2130,8 +2234,31 @@ class GraphExecutor:
                         )
 
                     if result.success:
-                        # Write outputs to shared memory using async write
+                        # Write outputs to shared memory with conflict detection
+                        conflict_strategy = self._parallel_config.memory_conflict_strategy
                         for key, value in result.output.items():
+                            async with fanout_keys_lock:
+                                prior_branch = fanout_written_keys.get(key)
+                                if prior_branch and prior_branch != branch.branch_id:
+                                    if conflict_strategy == "error":
+                                        raise RuntimeError(
+                                            f"Memory conflict: key '{key}' already written "
+                                            f"by branch '{prior_branch}', "
+                                            f"conflicting write from '{branch.branch_id}'"
+                                        )
+                                    elif conflict_strategy == "first_wins":
+                                        self.logger.debug(
+                                            f"      ⚠ Skipping write to '{key}' "
+                                            f"(first_wins: already set by {prior_branch})"
+                                        )
+                                        continue
+                                    else:
+                                        # last_wins (default): write and log
+                                        self.logger.debug(
+                                            f"      ⚠ Key '{key}' overwritten "
+                                            f"(last_wins: {prior_branch} -> {branch.branch_id})"
+                                        )
+                                fanout_written_keys[key] = branch.branch_id
                             await memory.write_async(key, value)
 
                         branch.result = result
@@ -2178,9 +2305,11 @@ class GraphExecutor:
 
                 return branch, e
 
-        # Execute all branches concurrently
-        tasks = [execute_single_branch(b) for b in branches.values()]
-        results = await asyncio.gather(*tasks, return_exceptions=False)
+        # Execute all branches concurrently with per-branch timeout
+        timeout = self._parallel_config.branch_timeout_seconds
+        branch_list = list(branches.values())
+        tasks = [asyncio.wait_for(execute_single_branch(b), timeout=timeout) for b in branch_list]
+        results = await asyncio.gather(*tasks, return_exceptions=True)
 
         # Process results
         total_tokens = 0
@@ -2188,17 +2317,33 @@ class GraphExecutor:
         branch_results: dict[str, NodeResult] = {}
         failed_branches: list[ParallelBranch] = []
 
-        for branch, result in results:
-            path.append(branch.node_id)
+        for i, result in enumerate(results):
+            branch = branch_list[i]
 
-            if isinstance(result, Exception):
+            if isinstance(result, asyncio.TimeoutError):
+                # Branch timed out
+                branch.status = "timed_out"
+                branch.error = f"Branch timed out after {timeout}s"
+                self.logger.warning(
+                    f"      ⏱ Branch {graph.get_node(branch.node_id).name}: "
+                    f"timed out after {timeout}s"
+                )
+                path.append(branch.node_id)
                 failed_branches.append(branch)
-            elif result is None or not result.success:
+            elif isinstance(result, Exception):
+                path.append(branch.node_id)
                 failed_branches.append(branch)
             else:
-                total_tokens += result.tokens_used
-                total_latency += result.latency_ms
-                branch_results[branch.branch_id] = result
+                returned_branch, node_result = result
+                path.append(returned_branch.node_id)
+                if node_result is None or isinstance(node_result, Exception):
+                    failed_branches.append(returned_branch)
+                elif not node_result.success:
+                    failed_branches.append(returned_branch)
+                else:
+                    total_tokens += node_result.tokens_used
+                    total_latency += node_result.latency_ms
+                    branch_results[returned_branch.branch_id] = node_result
 
         # Handle failures based on config
         if failed_branches:

core/framework/graph/gcu.py

@@ -37,24 +37,42 @@ Follow these rules for reliable, efficient browser interaction.
 ## Reading Pages
 - ALWAYS prefer `browser_snapshot` over `browser_get_text("body")`
   — it returns a compact ~1-5 KB accessibility tree vs 100+ KB of raw HTML.
-- Use `browser_snapshot_aria` when you need full ARIA properties
-  for detailed element inspection.
+- Interaction tools (`browser_click`, `browser_type`, `browser_fill`,
+  `browser_scroll`, etc.) return a page snapshot automatically in their
+  result. Use it to decide your next action — do NOT call
+  `browser_snapshot` separately after every action.
+  Only call `browser_snapshot` when you need a fresh view without
+  performing an action, or after setting `auto_snapshot=false`.
 - Do NOT use `browser_screenshot` for reading text content
   — it produces huge base64 images with no searchable text.
 - Only fall back to `browser_get_text` for extracting specific
   small elements by CSS selector.
 
 ## Navigation & Waiting
-- Always call `browser_wait` after navigation actions
-  (`browser_open`, `browser_navigate`, `browser_click` on links)
-  to let the page load.
+- `browser_navigate` and `browser_open` already wait for the page to
+  load (`domcontentloaded`). Do NOT call `browser_wait` with no
+  arguments after navigation — it wastes time.
+  Only use `browser_wait` when you need a *specific element* or *text*
+  to appear (pass `selector` or `text`).
 - NEVER re-navigate to the same URL after scrolling
   — this resets your scroll position and loses loaded content.
 
 ## Scrolling
 - Use large scroll amounts ~2000 when loading more content
   — sites like twitter and linkedin have lazy loading for paging.
-- After scrolling, take a new `browser_snapshot` to see updated content.
+- The scroll result includes a snapshot automatically — no need to call
+  `browser_snapshot` separately.
+
+## Batching Actions
+- You can call multiple tools in a single turn — they execute in parallel.
+  ALWAYS batch independent actions together. Examples:
+  - Fill multiple form fields in one turn.
+  - Navigate + snapshot in one turn.
+  - Click + scroll if targeting different elements.
+- When batching, set `auto_snapshot=false` on all but the last action
+  to avoid redundant snapshots.
+- Aim for 3-5 tool calls per turn minimum. One tool call per turn is
+  wasteful.
 
 ## Error Recovery
 - If a tool fails, retry once with the same approach.
@@ -65,11 +83,33 @@ Follow these rules for reliable, efficient browser interaction.
   then `browser_start`, then retry.
 
 ## Tab Management
-- Use `browser_tabs` to list open tabs when managing multiple pages.
-- Pass `target_id` to tools when operating on a specific tab.
-- Open background tabs with `browser_open(url=..., background=true)`
-  to avoid losing your current context.
-- Close tabs you no longer need with `browser_close` to free resources.
+
+**Close tabs as soon as you are done with them** — not only at the end of the task.
+After reading or extracting data from a tab, close it immediately.
+
+**Decision rules:**
+- Finished reading/extracting from a tab? → `browser_close(target_id=...)`
+- Completed a multi-tab workflow? → `browser_close_finished()` to clean up all your tabs
+- More than 3 tabs open? → stop and close finished ones before opening more
+- Popup appeared that you didn't need? → close it immediately
+
+**Origin awareness:** `browser_tabs` returns an `origin` field for each tab:
+- `"agent"` — you opened it; you own it; close it when done
+- `"popup"` — opened by a link or script; close after extracting what you need
+- `"startup"` or `"user"` — leave these alone unless the task requires it
+
+**Cleanup tools:**
+- `browser_close(target_id=...)` — close one specific tab
+- `browser_close_finished()` — close all your agent/popup tabs (safe: leaves startup/user tabs)
+- `browser_close_all()` — close everything except the active tab (use only for full reset)
+
+**Multi-tab workflow pattern:**
+1. Open background tabs with `browser_open(url=..., background=true)` to stay on current tab
+2. Process each tab and close it with `browser_close` when done
+3. When the full workflow completes, call `browser_close_finished()` to confirm cleanup
+4. Check `browser_tabs` at any point — it shows `origin` and `age_seconds` per tab
+
+Never accumulate tabs. Treat every tab you open as a resource you must free.
 
 ## Login & Auth Walls
 - If you see a "Log in" or "Sign up" prompt instead of expected

core/framework/graph/node.py

@@ -565,6 +565,15 @@ class NodeContext:
     # staging / running) without restarting the conversation.
     dynamic_prompt_provider: Any = None  # Callable[[], str] | None
 
+    # Skill system prompts — injected by the skill discovery pipeline
+    skills_catalog_prompt: str = ""  # Available skills XML catalog
+    protocols_prompt: str = ""  # Default skill operational protocols
+
+    # Per-iteration metadata provider — when set, EventLoopNode merges
+    # the returned dict into node_loop_iteration event data.  Used by
+    # the queen to record the current phase per iteration.
+    iteration_metadata_provider: Any = None  # Callable[[], dict] | None
+
 
 @dataclass
 class NodeResult:

core/framework/graph/prompt_composer.py

@@ -26,6 +26,16 @@ if TYPE_CHECKING:
 
 logger = logging.getLogger(__name__)
 
+# Injected into every worker node's system prompt so the LLM understands
+# it is one step in a multi-node pipeline and should not overreach.
+EXECUTION_SCOPE_PREAMBLE = (
+    "EXECUTION SCOPE: You are one node in a multi-step workflow graph. "
+    "Focus ONLY on the task described in your instructions below. "
+    "Call set_output() for each of your declared output keys, then stop. "
+    "Do NOT attempt work that belongs to other nodes — the framework "
+    "routes data between nodes automatically."
+)
+
 
 def _with_datetime(prompt: str) -> str:
     """Append current datetime with local timezone to a system prompt."""
@@ -140,14 +150,18 @@ def compose_system_prompt(
     focus_prompt: str | None,
     narrative: str | None = None,
     accounts_prompt: str | None = None,
+    skills_catalog_prompt: str | None = None,
+    protocols_prompt: str | None = None,
 ) -> str:
-    """Compose the three-layer system prompt.
+    """Compose the multi-layer system prompt.
 
     Args:
         identity_prompt: Layer 1 — static agent identity (from GraphSpec).
         focus_prompt: Layer 3 — per-node focus directive (from NodeSpec.system_prompt).
         narrative: Layer 2 — auto-generated from conversation state.
         accounts_prompt: Connected accounts block (sits between identity and narrative).
+        skills_catalog_prompt: Available skills catalog XML (Agent Skills standard).
+        protocols_prompt: Default skill operational protocols section.
 
     Returns:
         Composed system prompt with all layers present, plus current datetime.
@@ -162,6 +176,14 @@ def compose_system_prompt(
     if accounts_prompt:
         parts.append(f"\n{accounts_prompt}")
 
+    # Skills catalog (discovered skills available for activation)
+    if skills_catalog_prompt:
+        parts.append(f"\n{skills_catalog_prompt}")
+
+    # Operational protocols (default skill behavioral guidance)
+    if protocols_prompt:
+        parts.append(f"\n{protocols_prompt}")
+
     # Layer 2: Narrative (what's happened so far)
     if narrative:
         parts.append(f"\n--- Context (what has happened so far) ---\n{narrative}")
@@ -255,7 +277,9 @@ def build_transition_marker(
     sections.append(f"\nCompleted: {previous_node.name}")
     sections.append(f"  {previous_node.description}")
 
-    # Outputs in memory
+    # Outputs in memory — use file references for large values so the
+    # next node loads full data from disk instead of seeing truncated
+    # inline previews that look deceptively complete.
     all_memory = memory.read_all()
     if all_memory:
         memory_lines: list[str] = []
@@ -263,7 +287,29 @@ def build_transition_marker(
             if value is None:
                 continue
             val_str = str(value)
-            if len(val_str) > 300:
+            if len(val_str) > 300 and data_dir:
+                # Auto-spill large transition values to data files
+                import json as _json
+
+                data_path = Path(data_dir)
+                data_path.mkdir(parents=True, exist_ok=True)
+                ext = ".json" if isinstance(value, (dict, list)) else ".txt"
+                filename = f"output_{key}{ext}"
+                try:
+                    write_content = (
+                        _json.dumps(value, indent=2, ensure_ascii=False)
+                        if isinstance(value, (dict, list))
+                        else str(value)
+                    )
+                    (data_path / filename).write_text(write_content, encoding="utf-8")
+                    file_size = (data_path / filename).stat().st_size
+                    val_str = (
+                        f"[Saved to '{filename}' ({file_size:,} bytes). "
+                        f"Use load_data(filename='{filename}') to access.]"
+                    )
+                except Exception:
+                    val_str = val_str[:300] + "..."
+            elif len(val_str) > 300:
                 val_str = val_str[:300] + "..."
             memory_lines.append(f"  {key}: {val_str}")
         if memory_lines:
@@ -280,7 +326,7 @@ def build_transition_marker(
                 ]
                 if file_lines:
                     sections.append(
-                        "\nData files (use read_file to access):\n" + "\n".join(file_lines)
+                        "\nData files (use load_data to access):\n" + "\n".join(file_lines)
                     )
 
     # Agent working memory
@@ -294,6 +340,12 @@ def build_transition_marker(
     # Next phase
     sections.append(f"\nNow entering: {next_node.name}")
     sections.append(f"  {next_node.description}")
+    if next_node.output_keys:
+        sections.append(
+            f"\nYour ONLY job in this phase: complete the task above and call "
+            f"set_output() for {next_node.output_keys}. Do NOT do work that "
+            f"belongs to later phases."
+        )
 
     # Reflection prompt (engineered metacognition)
     sections.append(

core/framework/graph/safe_eval.py

@@ -115,11 +115,23 @@ class SafeEvalVisitor(ast.NodeVisitor):
         return True
 
     def visit_BoolOp(self, node: ast.BoolOp) -> Any:
-        values = [self.visit(v) for v in node.values]
+        # Short-circuit evaluation to match Python semantics.
+        # Previously all operands were eagerly evaluated, which broke
+        # guard patterns like: ``x is not None and x.get("key")``
         if isinstance(node.op, ast.And):
-            return all(values)
+            result = True
+            for v in node.values:
+                result = self.visit(v)
+                if not result:
+                    return result
+            return result
         elif isinstance(node.op, ast.Or):
-            return any(values)
+            result = False
+            for v in node.values:
+                result = self.visit(v)
+                if result:
+                    return result
+            return result
         raise ValueError(f"Boolean operator {type(node.op).__name__} is not allowed")
 
     def visit_IfExp(self, node: ast.IfExp) -> Any:

core/framework/llm/litellm.py

@@ -7,9 +7,11 @@ Groq, and local models.
 See: https://docs.litellm.ai/docs/providers
 """
 
+import ast
 import asyncio
 import json
 import logging
+import re
 import time
 from collections.abc import AsyncIterator
 from datetime import datetime
@@ -23,6 +25,7 @@ except ImportError:
     litellm = None  # type: ignore[assignment]
     RateLimitError = Exception  # type: ignore[assignment, misc]
 
+from framework.config import HIVE_LLM_ENDPOINT as HIVE_API_BASE
 from framework.llm.provider import LLMProvider, LLMResponse, Tool
 from framework.llm.stream_events import StreamEvent
 
@@ -45,6 +48,12 @@ def _patch_litellm_anthropic_oauth() -> None:
         from litellm.llms.anthropic.common_utils import AnthropicModelInfo
         from litellm.types.llms.anthropic import ANTHROPIC_OAUTH_TOKEN_PREFIX
     except ImportError:
+        logger.warning(
+            "Could not apply litellm Anthropic OAuth patch — litellm internals may have "
+            "changed. Anthropic OAuth tokens (Claude Code subscriptions) may fail with 401. "
+            "See BerriAI/litellm#19618. Current litellm version: %s",
+            getattr(litellm, "__version__", "unknown"),
+        )
         return
 
     original = AnthropicModelInfo.validate_environment
@@ -86,10 +95,12 @@ def _patch_litellm_metadata_nonetype() -> None:
     """
     import functools
 
+    patched_count = 0
     for fn_name in ("completion", "acompletion", "responses", "aresponses"):
         original = getattr(litellm, fn_name, None)
         if original is None:
             continue
+        patched_count += 1
         if asyncio.iscoroutinefunction(original):
 
             @functools.wraps(original)
@@ -109,6 +120,14 @@ def _patch_litellm_metadata_nonetype() -> None:
 
             setattr(litellm, fn_name, _sync_wrapper)
 
+    if patched_count == 0:
+        logger.warning(
+            "Could not apply litellm metadata=None patch — none of the expected entry "
+            "points (completion, acompletion, responses, aresponses) were found. "
+            "metadata=None TypeError may occur. Current litellm version: %s",
+            getattr(litellm, "__version__", "unknown"),
+        )
+
 
 if litellm is not None:
     _patch_litellm_anthropic_oauth()
@@ -118,6 +137,26 @@ RATE_LIMIT_MAX_RETRIES = 10
 RATE_LIMIT_BACKOFF_BASE = 2  # seconds
 RATE_LIMIT_MAX_DELAY = 120  # seconds - cap to prevent absurd waits
 MINIMAX_API_BASE = "https://api.minimax.io/v1"
+OPENROUTER_API_BASE = "https://openrouter.ai/api/v1"
+
+# Providers that accept cache_control on message content blocks.
+# Anthropic: native ephemeral caching. MiniMax & Z-AI/GLM: pass-through to their APIs.
+# (OpenAI caches automatically server-side; Groq/Gemini/etc. strip the header.)
+_CACHE_CONTROL_PREFIXES = (
+    "anthropic/",
+    "claude-",
+    "minimax/",
+    "minimax-",
+    "MiniMax-",
+    "zai-glm",
+    "glm-",
+)
+
+
+def _model_supports_cache_control(model: str) -> bool:
+    return any(model.startswith(p) for p in _CACHE_CONTROL_PREFIXES)
+
+
 # Kimi For Coding uses an Anthropic-compatible endpoint (no /v1 suffix).
 # Claude Code integration uses this format; the /v1 OpenAI-compatible endpoint
 # enforces a coding-agent whitelist that blocks unknown User-Agents.
@@ -127,10 +166,26 @@ KIMI_API_BASE = "https://api.kimi.com/coding"
 # Conversation-structure issues are deterministic — long waits don't help.
 EMPTY_STREAM_MAX_RETRIES = 3
 EMPTY_STREAM_RETRY_DELAY = 1.0  # seconds
+OPENROUTER_TOOL_COMPAT_ERROR_SNIPPETS = (
+    "no endpoints found that support tool use",
+    "no endpoints available that support tool use",
+    "provider routing",
+)
+OPENROUTER_TOOL_CALL_RE = re.compile(
+    r"<\|tool_call_start\|>\s*(.*?)\s*<\|tool_call_end\|>",
+    re.DOTALL,
+)
+OPENROUTER_TOOL_COMPAT_CACHE_TTL_SECONDS = 3600
+# OpenRouter routing can change over time, so tool-compat caching must expire.
+OPENROUTER_TOOL_COMPAT_MODEL_CACHE: dict[str, float] = {}
 
 # Directory for dumping failed requests
 FAILED_REQUESTS_DIR = Path.home() / ".hive" / "failed_requests"
 
+# Maximum number of dump files to retain in ~/.hive/failed_requests/.
+# Older files are pruned automatically to prevent unbounded disk growth.
+MAX_FAILED_REQUEST_DUMPS = 50
+
 
 def _estimate_tokens(model: str, messages: list[dict]) -> tuple[int, str]:
     """Estimate token count for messages. Returns (token_count, method)."""
@@ -147,6 +202,42 @@ def _estimate_tokens(model: str, messages: list[dict]) -> tuple[int, str]:
     return total_chars // 4, "estimate"
 
 
+def _prune_failed_request_dumps(max_files: int = MAX_FAILED_REQUEST_DUMPS) -> None:
+    """Remove oldest dump files when the count exceeds *max_files*.
+
+    Best-effort: never raises — a pruning failure must not break retry logic.
+    """
+    try:
+        all_dumps = sorted(
+            FAILED_REQUESTS_DIR.glob("*.json"),
+            key=lambda f: f.stat().st_mtime,
+        )
+        excess = len(all_dumps) - max_files
+        if excess > 0:
+            for old_file in all_dumps[:excess]:
+                old_file.unlink(missing_ok=True)
+    except Exception:
+        pass  # Best-effort — never block the caller
+
+
+def _remember_openrouter_tool_compat_model(model: str) -> None:
+    """Cache OpenRouter tool-compat fallback for a bounded time window."""
+    OPENROUTER_TOOL_COMPAT_MODEL_CACHE[model] = (
+        time.monotonic() + OPENROUTER_TOOL_COMPAT_CACHE_TTL_SECONDS
+    )
+
+
+def _is_openrouter_tool_compat_cached(model: str) -> bool:
+    """Return True when the cached OpenRouter compat entry is still fresh."""
+    expires_at = OPENROUTER_TOOL_COMPAT_MODEL_CACHE.get(model)
+    if expires_at is None:
+        return False
+    if expires_at <= time.monotonic():
+        OPENROUTER_TOOL_COMPAT_MODEL_CACHE.pop(model, None)
+        return False
+    return True
+
+
 def _dump_failed_request(
     model: str,
     kwargs: dict[str, Any],
@@ -178,6 +269,9 @@ def _dump_failed_request(
     with open(filepath, "w", encoding="utf-8") as f:
         json.dump(dump_data, f, indent=2, default=str)
 
+    # Prune old dumps to prevent unbounded disk growth
+    _prune_failed_request_dumps()
+
     return str(filepath)
 
 
@@ -339,6 +433,10 @@ class LiteLLMProvider(LLMProvider):
             # Strip a trailing /v1 in case the user's saved config has the old value.
             if api_base and api_base.rstrip("/").endswith("/v1"):
                 api_base = api_base.rstrip("/")[:-3]
+        elif model.lower().startswith("hive/"):
+            model = "anthropic/" + model[len("hive/") :]
+            if api_base and api_base.rstrip("/").endswith("/v1"):
+                api_base = api_base.rstrip("/")[:-3]
         self.model = model
         self.api_key = api_key
         self.api_base = api_base or self._default_api_base_for_model(_original_model)
@@ -366,8 +464,12 @@ class LiteLLMProvider(LLMProvider):
         model_lower = model.lower()
         if model_lower.startswith("minimax/") or model_lower.startswith("minimax-"):
             return MINIMAX_API_BASE
+        if model_lower.startswith("openrouter/"):
+            return OPENROUTER_API_BASE
         if model_lower.startswith("kimi/"):
             return KIMI_API_BASE
+        if model_lower.startswith("hive/"):
+            return HIVE_API_BASE
         return None
 
     def _completion_with_rate_limit_retry(
@@ -707,7 +809,10 @@ class LiteLLMProvider(LLMProvider):
 
         full_messages: list[dict[str, Any]] = []
         if system:
-            full_messages.append({"role": "system", "content": system})
+            sys_msg: dict[str, Any] = {"role": "system", "content": system}
+            if _model_supports_cache_control(self.model):
+                sys_msg["cache_control"] = {"type": "ephemeral"}
+            full_messages.append(sys_msg)
         full_messages.extend(messages)
 
         if json_mode:
@@ -769,6 +874,494 @@ class LiteLLMProvider(LLMProvider):
         model = (self.model or "").lower()
         return model.startswith("minimax/") or model.startswith("minimax-")
 
+    def _is_openrouter_model(self) -> bool:
+        """Return True when the configured model targets OpenRouter."""
+        model = (self.model or "").lower()
+        if model.startswith("openrouter/"):
+            return True
+        api_base = (self.api_base or "").lower()
+        return "openrouter.ai/api/v1" in api_base
+
+    def _should_use_openrouter_tool_compat(
+        self,
+        error: BaseException,
+        tools: list[Tool] | None,
+    ) -> bool:
+        """Return True when OpenRouter rejects native tool use for the model."""
+        if not tools or not self._is_openrouter_model():
+            return False
+        error_text = str(error).lower()
+        return "openrouter" in error_text and any(
+            snippet in error_text for snippet in OPENROUTER_TOOL_COMPAT_ERROR_SNIPPETS
+        )
+
+    @staticmethod
+    def _extract_json_object(text: str) -> dict[str, Any] | None:
+        """Extract the first JSON object from a model response."""
+        candidates = [text.strip()]
+
+        stripped = text.strip()
+        if stripped.startswith("```"):
+            fence_lines = stripped.splitlines()
+            if len(fence_lines) >= 3:
+                candidates.append("\n".join(fence_lines[1:-1]).strip())
+
+        decoder = json.JSONDecoder()
+        for candidate in candidates:
+            if not candidate:
+                continue
+            try:
+                parsed = json.loads(candidate)
+            except json.JSONDecodeError:
+                parsed = None
+            if isinstance(parsed, dict):
+                return parsed
+
+            for start_idx, char in enumerate(candidate):
+                if char != "{":
+                    continue
+                try:
+                    parsed, _ = decoder.raw_decode(candidate[start_idx:])
+                except json.JSONDecodeError:
+                    continue
+                if isinstance(parsed, dict):
+                    return parsed
+        return None
+
+    def _parse_openrouter_tool_compat_response(
+        self,
+        content: str,
+        tools: list[Tool],
+    ) -> tuple[str, list[dict[str, Any]]]:
+        """Parse JSON tool-compat output into assistant text and tool calls."""
+        payload = self._extract_json_object(content)
+        if payload is None:
+            text_tool_content, text_tool_calls = self._parse_openrouter_text_tool_calls(
+                content,
+                tools,
+            )
+            if text_tool_calls:
+                logger.info(
+                    "[openrouter-tool-compat] Parsed textual tool-call markers for %s",
+                    self.model,
+                )
+                return text_tool_content, text_tool_calls
+            logger.info(
+                "[openrouter-tool-compat] %s returned non-JSON fallback content; "
+                "treating it as plain text.",
+                self.model,
+            )
+            return content.strip(), []
+
+        assistant_text = payload.get("assistant_response")
+        if not isinstance(assistant_text, str):
+            assistant_text = payload.get("content")
+        if not isinstance(assistant_text, str):
+            assistant_text = payload.get("response")
+        if not isinstance(assistant_text, str):
+            assistant_text = ""
+
+        tool_calls_raw = payload.get("tool_calls")
+        if not tool_calls_raw and {"name", "arguments"} <= payload.keys():
+            tool_calls_raw = [payload]
+        elif isinstance(payload.get("tool_call"), dict):
+            tool_calls_raw = [payload["tool_call"]]
+
+        if not isinstance(tool_calls_raw, list):
+            tool_calls_raw = []
+
+        allowed_tool_names = {tool.name for tool in tools}
+        tool_calls: list[dict[str, Any]] = []
+        compat_prefix = f"openrouter_compat_{time.time_ns()}"
+
+        for idx, raw_call in enumerate(tool_calls_raw):
+            if not isinstance(raw_call, dict):
+                continue
+
+            function_block = raw_call.get("function")
+            function_name = (
+                raw_call.get("name")
+                or raw_call.get("tool_name")
+                or (function_block.get("name") if isinstance(function_block, dict) else None)
+            )
+            if not isinstance(function_name, str) or function_name not in allowed_tool_names:
+                if function_name:
+                    logger.warning(
+                        "[openrouter-tool-compat] Ignoring unknown tool '%s' for model %s",
+                        function_name,
+                        self.model,
+                    )
+                continue
+
+            arguments = raw_call.get("arguments")
+            if arguments is None:
+                arguments = raw_call.get("tool_input")
+            if arguments is None:
+                arguments = raw_call.get("input")
+            if arguments is None and isinstance(function_block, dict):
+                arguments = function_block.get("arguments")
+            if arguments is None:
+                arguments = {}
+
+            if isinstance(arguments, str):
+                try:
+                    arguments = json.loads(arguments)
+                except json.JSONDecodeError:
+                    arguments = {"_raw": arguments}
+            elif not isinstance(arguments, dict):
+                arguments = {"value": arguments}
+
+            tool_calls.append(
+                {
+                    "id": f"{compat_prefix}_{idx}",
+                    "name": function_name,
+                    "input": arguments,
+                }
+            )
+
+        return assistant_text.strip(), tool_calls
+
+    @staticmethod
+    def _close_truncated_json_fragment(fragment: str) -> str:
+        """Close a truncated JSON fragment by balancing quotes/brackets."""
+        stack: list[str] = []
+        in_string = False
+        escaped = False
+        normalized = fragment.rstrip()
+
+        while normalized and normalized[-1] in ",:{[":
+            normalized = normalized[:-1].rstrip()
+
+        for char in normalized:
+            if in_string:
+                if escaped:
+                    escaped = False
+                elif char == "\\":
+                    escaped = True
+                elif char == '"':
+                    in_string = False
+                continue
+
+            if char == '"':
+                in_string = True
+            elif char in "{[":
+                stack.append(char)
+            elif char == "}" and stack and stack[-1] == "{":
+                stack.pop()
+            elif char == "]" and stack and stack[-1] == "[":
+                stack.pop()
+
+        if in_string:
+            if escaped:
+                normalized = normalized[:-1]
+            normalized += '"'
+
+        for opener in reversed(stack):
+            normalized += "}" if opener == "{" else "]"
+
+        return normalized
+
+    def _repair_truncated_tool_arguments(self, raw_arguments: str) -> dict[str, Any] | None:
+        """Try to recover a truncated JSON object from tool-call arguments."""
+        stripped = raw_arguments.strip()
+        if not stripped or stripped[0] != "{":
+            return None
+
+        max_trim = min(len(stripped), 256)
+        for trim in range(max_trim + 1):
+            candidate = stripped[: len(stripped) - trim].rstrip()
+            if not candidate:
+                break
+            candidate = self._close_truncated_json_fragment(candidate)
+            try:
+                parsed = json.loads(candidate)
+            except json.JSONDecodeError:
+                continue
+            if isinstance(parsed, dict):
+                return parsed
+        return None
+
+    def _parse_tool_call_arguments(self, raw_arguments: str, tool_name: str) -> dict[str, Any]:
+        """Parse streamed tool arguments, repairing truncation when possible."""
+        try:
+            parsed = json.loads(raw_arguments) if raw_arguments else {}
+        except json.JSONDecodeError:
+            parsed = None
+
+        if isinstance(parsed, dict):
+            return parsed
+
+        repaired = self._repair_truncated_tool_arguments(raw_arguments)
+        if repaired is not None:
+            logger.warning(
+                "[tool-args] Recovered truncated arguments for %s on %s",
+                tool_name,
+                self.model,
+            )
+            return repaired
+
+        raise ValueError(
+            f"Failed to parse tool call arguments for '{tool_name}' (likely truncated JSON)."
+        )
+
+    def _parse_openrouter_text_tool_calls(
+        self,
+        content: str,
+        tools: list[Tool],
+    ) -> tuple[str, list[dict[str, Any]]]:
+        """Parse textual OpenRouter tool calls into synthetic tool calls.
+
+        Supports both:
+        - Marker wrapped payloads: <|tool_call_start|>...<|tool_call_end|>
+        - Plain one-line tool calls: ask_user("...", ["..."])
+        """
+        tools_by_name = {tool.name: tool for tool in tools}
+        compat_prefix = f"openrouter_compat_{time.time_ns()}"
+        tool_calls: list[dict[str, Any]] = []
+        segment_index = 0
+
+        for match in OPENROUTER_TOOL_CALL_RE.finditer(content):
+            parsed_calls = self._parse_openrouter_text_tool_call_block(
+                block=match.group(1),
+                tools_by_name=tools_by_name,
+                compat_prefix=f"{compat_prefix}_{segment_index}",
+            )
+            if parsed_calls:
+                segment_index += 1
+                tool_calls.extend(parsed_calls)
+
+        stripped_content = OPENROUTER_TOOL_CALL_RE.sub("", content)
+        retained_lines: list[str] = []
+        for line in stripped_content.splitlines():
+            stripped_line = line.strip()
+            if not stripped_line:
+                retained_lines.append(line)
+                continue
+
+            candidate = stripped_line
+            if candidate.startswith("`") and candidate.endswith("`") and len(candidate) > 1:
+                candidate = candidate[1:-1].strip()
+
+            parsed_calls = self._parse_openrouter_text_tool_call_block(
+                block=candidate,
+                tools_by_name=tools_by_name,
+                compat_prefix=f"{compat_prefix}_{segment_index}",
+            )
+            if parsed_calls:
+                segment_index += 1
+                tool_calls.extend(parsed_calls)
+                continue
+
+            retained_lines.append(line)
+
+        stripped_text = "\n".join(retained_lines).strip()
+        return stripped_text, tool_calls
+
+    def _parse_openrouter_text_tool_call_block(
+        self,
+        block: str,
+        tools_by_name: dict[str, Tool],
+        compat_prefix: str,
+    ) -> list[dict[str, Any]]:
+        """Parse a single textual tool-call block like [tool(arg='x')]."""
+        try:
+            parsed = ast.parse(block.strip(), mode="eval").body
+        except SyntaxError:
+            return []
+
+        call_nodes = parsed.elts if isinstance(parsed, ast.List) else [parsed]
+        tool_calls: list[dict[str, Any]] = []
+
+        for call_index, call_node in enumerate(call_nodes):
+            if not isinstance(call_node, ast.Call) or not isinstance(call_node.func, ast.Name):
+                continue
+
+            tool_name = call_node.func.id
+            tool = tools_by_name.get(tool_name)
+            if tool is None:
+                continue
+
+            try:
+                tool_input = self._parse_openrouter_text_tool_call_arguments(
+                    call_node=call_node,
+                    tool=tool,
+                )
+            except (ValueError, SyntaxError):
+                continue
+
+            tool_calls.append(
+                {
+                    "id": f"{compat_prefix}_{call_index}",
+                    "name": tool_name,
+                    "input": tool_input,
+                }
+            )
+
+        return tool_calls
+
+    @staticmethod
+    def _parse_openrouter_text_tool_call_arguments(
+        call_node: ast.Call,
+        tool: Tool,
+    ) -> dict[str, Any]:
+        """Parse positional/keyword args from a textual tool call."""
+        properties = tool.parameters.get("properties", {})
+        positional_keys = list(properties.keys())
+        tool_input: dict[str, Any] = {}
+
+        if len(call_node.args) > len(positional_keys):
+            raise ValueError("Too many positional args for textual tool call")
+
+        for idx, arg_node in enumerate(call_node.args):
+            tool_input[positional_keys[idx]] = ast.literal_eval(arg_node)
+
+        for kwarg in call_node.keywords:
+            if kwarg.arg is None:
+                raise ValueError("Star args are not supported in textual tool calls")
+            tool_input[kwarg.arg] = ast.literal_eval(kwarg.value)
+
+        return tool_input
+
+    def _build_openrouter_tool_compat_messages(
+        self,
+        messages: list[dict[str, Any]],
+        system: str,
+        tools: list[Tool],
+    ) -> list[dict[str, Any]]:
+        """Build a JSON-only prompt for models without native tool support."""
+        tool_specs = [
+            {
+                "name": tool.name,
+                "description": tool.description,
+                "parameters": tool.parameters,
+            }
+            for tool in tools
+        ]
+        compat_instruction = (
+            "Tool compatibility mode is active because this OpenRouter model does not support "
+            "native function calling on the routed provider.\n"
+            "Return exactly one JSON object and nothing else.\n"
+            'Schema: {"assistant_response": string, '
+            '"tool_calls": [{"name": string, "arguments": object}]}\n'
+            "Rules:\n"
+            "- If a tool is required, put one or more entries in tool_calls "
+            "and do not invent tool results.\n"
+            "- If no tool is required, set tool_calls to [] and put the full "
+            "answer in assistant_response.\n"
+            "- Only use tool names from the allowed tool list.\n"
+            "- arguments must always be valid JSON objects.\n"
+            f"Allowed tools:\n{json.dumps(tool_specs, ensure_ascii=True)}"
+        )
+        compat_system = compat_instruction if not system else f"{system}\n\n{compat_instruction}"
+
+        full_messages: list[dict[str, Any]] = [{"role": "system", "content": compat_system}]
+        full_messages.extend(messages)
+        return [
+            message
+            for message in full_messages
+            if not (
+                message.get("role") == "assistant"
+                and not message.get("content")
+                and not message.get("tool_calls")
+            )
+        ]
+
+    async def _acomplete_via_openrouter_tool_compat(
+        self,
+        messages: list[dict[str, Any]],
+        system: str,
+        tools: list[Tool],
+        max_tokens: int,
+    ) -> LLMResponse:
+        """Emulate tool calling via JSON when OpenRouter rejects native tools."""
+        full_messages = self._build_openrouter_tool_compat_messages(messages, system, tools)
+        kwargs: dict[str, Any] = {
+            "model": self.model,
+            "messages": full_messages,
+            "max_tokens": max_tokens,
+            **self.extra_kwargs,
+        }
+        if self.api_key:
+            kwargs["api_key"] = self.api_key
+        if self.api_base:
+            kwargs["api_base"] = self.api_base
+
+        response = await self._acompletion_with_rate_limit_retry(**kwargs)
+        raw_content = response.choices[0].message.content or ""
+        assistant_text, tool_calls = self._parse_openrouter_tool_compat_response(
+            raw_content,
+            tools,
+        )
+        usage = response.usage
+        input_tokens = usage.prompt_tokens if usage else 0
+        output_tokens = usage.completion_tokens if usage else 0
+        stop_reason = "tool_calls" if tool_calls else (response.choices[0].finish_reason or "stop")
+
+        return LLMResponse(
+            content=assistant_text,
+            model=response.model or self.model,
+            input_tokens=input_tokens,
+            output_tokens=output_tokens,
+            stop_reason=stop_reason,
+            raw_response={
+                "compat_mode": "openrouter_tool_emulation",
+                "tool_calls": tool_calls,
+                "response": response,
+            },
+        )
+
+    async def _stream_via_openrouter_tool_compat(
+        self,
+        messages: list[dict[str, Any]],
+        system: str,
+        tools: list[Tool],
+        max_tokens: int,
+    ) -> AsyncIterator[StreamEvent]:
+        """Fallback stream for OpenRouter models without native tool support."""
+        from framework.llm.stream_events import (
+            FinishEvent,
+            StreamErrorEvent,
+            TextDeltaEvent,
+            TextEndEvent,
+            ToolCallEvent,
+        )
+
+        logger.info(
+            "[openrouter-tool-compat] Using compatibility mode for %s",
+            self.model,
+        )
+        try:
+            response = await self._acomplete_via_openrouter_tool_compat(
+                messages=messages,
+                system=system,
+                tools=tools,
+                max_tokens=max_tokens,
+            )
+        except Exception as e:
+            yield StreamErrorEvent(error=str(e), recoverable=False)
+            return
+
+        raw_response = response.raw_response if isinstance(response.raw_response, dict) else {}
+        tool_calls = raw_response.get("tool_calls", [])
+
+        if response.content:
+            yield TextDeltaEvent(content=response.content, snapshot=response.content)
+            yield TextEndEvent(full_text=response.content)
+
+        for tool_call in tool_calls:
+            yield ToolCallEvent(
+                tool_use_id=tool_call["id"],
+                tool_name=tool_call["name"],
+                tool_input=tool_call["input"],
+            )
+
+        yield FinishEvent(
+            stop_reason=response.stop_reason,
+            input_tokens=response.input_tokens,
+            output_tokens=response.output_tokens,
+            model=response.model,
+        )
+
     async def _stream_via_nonstream_completion(
         self,
         messages: list[dict[str, Any]],
@@ -812,12 +1405,11 @@ class LiteLLMProvider(LLMProvider):
             tool_calls = msg.tool_calls or []
 
         for tc in tool_calls:
-            parsed_args: Any
             args = tc.function.arguments if tc.function else ""
-            try:
-                parsed_args = json.loads(args) if args else {}
-            except json.JSONDecodeError:
-                parsed_args = {"_raw": args}
+            parsed_args = self._parse_tool_call_arguments(
+                args,
+                tc.function.name if tc.function else "",
+            )
             yield ToolCallEvent(
                 tool_use_id=getattr(tc, "id", ""),
                 tool_name=tc.function.name if tc.function else "",
@@ -876,9 +1468,22 @@ class LiteLLMProvider(LLMProvider):
                 yield event
             return
 
+        if tools and self._is_openrouter_model() and _is_openrouter_tool_compat_cached(self.model):
+            async for event in self._stream_via_openrouter_tool_compat(
+                messages=messages,
+                system=system,
+                tools=tools,
+                max_tokens=max_tokens,
+            ):
+                yield event
+            return
+
         full_messages: list[dict[str, Any]] = []
         if system:
-            full_messages.append({"role": "system", "content": system})
+            sys_msg: dict[str, Any] = {"role": "system", "content": system}
+            if _model_supports_cache_control(self.model):
+                sys_msg["cache_control"] = {"type": "ephemeral"}
+            full_messages.append(sys_msg)
         full_messages.extend(messages)
 
         # Codex Responses API requires an `instructions` field (system prompt).
@@ -943,9 +1548,26 @@ class LiteLLMProvider(LLMProvider):
                 response = await litellm.acompletion(**kwargs)  # type: ignore[union-attr]
 
                 async for chunk in response:
-                    choice = chunk.choices[0] if chunk.choices else None
-                    if not choice:
+                    # Capture usage from the trailing usage-only chunk that
+                    # stream_options={"include_usage": True} sends with empty choices.
+                    if not chunk.choices:
+                        usage = getattr(chunk, "usage", None)
+                        if usage:
+                            input_tokens = getattr(usage, "prompt_tokens", 0) or 0
+                            output_tokens = getattr(usage, "completion_tokens", 0) or 0
+                            logger.debug(
+                                "[tokens] trailing usage chunk: input=%d output=%d model=%s",
+                                input_tokens,
+                                output_tokens,
+                                self.model,
+                            )
+                        else:
+                            logger.debug(
+                                "[tokens] empty-choices chunk with no usage (model=%s)",
+                                self.model,
+                            )
                         continue
+                    choice = chunk.choices[0]
 
                     delta = choice.delta
 
@@ -1002,10 +1624,10 @@ class LiteLLMProvider(LLMProvider):
                     if choice.finish_reason:
                         stream_finish_reason = choice.finish_reason
                         for _idx, tc_data in sorted(tool_calls_acc.items()):
-                            try:
-                                parsed_args = json.loads(tc_data["arguments"])
-                            except (json.JSONDecodeError, KeyError):
-                                parsed_args = {"_raw": tc_data.get("arguments", "")}
+                            parsed_args = self._parse_tool_call_arguments(
+                                tc_data.get("arguments", ""),
+                                tc_data.get("name", ""),
+                            )
                             tail_events.append(
                                 ToolCallEvent(
                                     tool_use_id=tc_data["id"],
@@ -1018,19 +1640,91 @@ class LiteLLMProvider(LLMProvider):
                             tail_events.append(TextEndEvent(full_text=accumulated_text))
 
                         usage = getattr(chunk, "usage", None)
+                        logger.debug(
+                            "[tokens] finish-chunk raw usage: %r (type=%s)",
+                            usage,
+                            type(usage).__name__,
+                        )
+                        cached_tokens = 0
                         if usage:
                             input_tokens = getattr(usage, "prompt_tokens", 0) or 0
                             output_tokens = getattr(usage, "completion_tokens", 0) or 0
+                            _details = getattr(usage, "prompt_tokens_details", None)
+                            cached_tokens = (
+                                getattr(_details, "cached_tokens", 0) or 0
+                                if _details is not None
+                                else getattr(usage, "cache_read_input_tokens", 0) or 0
+                            )
+                            logger.debug(
+                                "[tokens] finish-chunk usage: "
+                                "input=%d output=%d cached=%d model=%s",
+                                input_tokens,
+                                output_tokens,
+                                cached_tokens,
+                                self.model,
+                            )
 
+                        logger.debug(
+                            "[tokens] finish event: input=%d output=%d cached=%d stop=%s model=%s",
+                            input_tokens,
+                            output_tokens,
+                            cached_tokens,
+                            choice.finish_reason,
+                            self.model,
+                        )
                         tail_events.append(
                             FinishEvent(
                                 stop_reason=choice.finish_reason,
                                 input_tokens=input_tokens,
                                 output_tokens=output_tokens,
+                                cached_tokens=cached_tokens,
                                 model=self.model,
                             )
                         )
 
+                # Fallback: LiteLLM strips usage from yielded chunks before
+                # returning them to us, but appends the original chunk (with
+                # usage intact) to response.chunks first.  Use LiteLLM's own
+                # calculate_total_usage() on that accumulated list.
+                if input_tokens == 0 and output_tokens == 0:
+                    try:
+                        from litellm.litellm_core_utils.streaming_handler import (
+                            calculate_total_usage,
+                        )
+
+                        _chunks = getattr(response, "chunks", None)
+                        if _chunks:
+                            _usage = calculate_total_usage(chunks=_chunks)
+                            input_tokens = _usage.prompt_tokens or 0
+                            output_tokens = _usage.completion_tokens or 0
+                            _details = getattr(_usage, "prompt_tokens_details", None)
+                            cached_tokens = (
+                                getattr(_details, "cached_tokens", 0) or 0
+                                if _details is not None
+                                else getattr(_usage, "cache_read_input_tokens", 0) or 0
+                            )
+                            logger.debug(
+                                "[tokens] post-loop chunks fallback:"
+                                " input=%d output=%d cached=%d model=%s",
+                                input_tokens,
+                                output_tokens,
+                                cached_tokens,
+                                self.model,
+                            )
+                            # Patch the FinishEvent already queued with 0 tokens
+                            for _i, _ev in enumerate(tail_events):
+                                if isinstance(_ev, FinishEvent) and _ev.input_tokens == 0:
+                                    tail_events[_i] = FinishEvent(
+                                        stop_reason=_ev.stop_reason,
+                                        input_tokens=input_tokens,
+                                        output_tokens=output_tokens,
+                                        cached_tokens=cached_tokens,
+                                        model=_ev.model,
+                                    )
+                                    break
+                    except Exception as _e:
+                        logger.debug("[tokens] chunks fallback failed: %s", _e)
+
                 # Check whether the stream produced any real content.
                 # (If text deltas were yielded above, has_content is True
                 # and we skip the retry path — nothing was yielded in vain.)
@@ -1114,6 +1808,16 @@ class LiteLLMProvider(LLMProvider):
                 return
 
             except Exception as e:
+                if self._should_use_openrouter_tool_compat(e, tools):
+                    _remember_openrouter_tool_compat_model(self.model)
+                    async for event in self._stream_via_openrouter_tool_compat(
+                        messages=messages,
+                        system=system,
+                        tools=tools or [],
+                        max_tokens=max_tokens,
+                    ):
+                        yield event
+                    return
                 if _is_stream_transient_error(e) and attempt < RATE_LIMIT_MAX_RETRIES:
                     wait = _compute_retry_delay(attempt, exception=e)
                     logger.warning(

core/framework/llm/stream_events.py

@@ -71,6 +71,7 @@ class FinishEvent:
     stop_reason: str = ""
     input_tokens: int = 0
     output_tokens: int = 0
+    cached_tokens: int = 0
     model: str = ""
 
 

core/framework/monitoring/__init__.py

@@ -1,33 +1 @@
-"""Framework-level worker monitoring package.
-
-Provides the Worker Health Judge: a reusable secondary graph that attaches to
-any worker agent runtime and monitors its execution health via periodic log
-inspection. Emits structured EscalationTickets when degradation is detected.
-
-Usage::
-
-    from framework.monitoring import HEALTH_JUDGE_ENTRY_POINT, judge_goal, judge_graph
-    from framework.tools.worker_monitoring_tools import register_worker_monitoring_tools
-
-    # Register tools bound to the worker runtime's EventBus
-    monitoring_registry = ToolRegistry()
-    register_worker_monitoring_tools(monitoring_registry, worker_runtime._event_bus, storage_path)
-
-    # Load judge as secondary graph on the worker runtime
-    await worker_runtime.add_graph(
-        graph_id="judge",
-        graph=judge_graph,
-        goal=judge_goal,
-        entry_points={"health_check": HEALTH_JUDGE_ENTRY_POINT},
-        storage_subpath="graphs/judge",
-    )
-"""
-
-from .judge import HEALTH_JUDGE_ENTRY_POINT, judge_goal, judge_graph, judge_node
-
-__all__ = [
-    "HEALTH_JUDGE_ENTRY_POINT",
-    "judge_goal",
-    "judge_graph",
-    "judge_node",
-]
+"""Framework-level worker monitoring package."""

core/framework/monitoring/judge.py

@@ -1,258 +0,0 @@
-"""Worker Health Judge — framework-level reusable monitoring graph.
-
-Attaches to any worker agent runtime as a secondary graph. Fires on a
-2-minute timer, reads the worker's session logs via ``get_worker_health_summary``,
-accumulates observations in a continuous conversation context, and emits a
-structured ``EscalationTicket`` when it detects a degradation pattern.
-
-Usage::
-
-    from framework.monitoring import judge_graph, judge_goal, HEALTH_JUDGE_ENTRY_POINT
-    from framework.tools.worker_monitoring_tools import register_worker_monitoring_tools
-
-    # Register tools bound to the worker runtime's event bus
-    monitoring_registry = ToolRegistry()
-    register_worker_monitoring_tools(
-        monitoring_registry, worker_runtime._event_bus, storage_path
-    )
-    monitoring_tools = list(monitoring_registry.get_tools().values())
-    monitoring_executor = monitoring_registry.get_executor()
-
-    # Load judge as secondary graph on the worker runtime
-    await worker_runtime.add_graph(
-        graph_id="judge",
-        graph=judge_graph,
-        goal=judge_goal,
-        entry_points={"health_check": HEALTH_JUDGE_ENTRY_POINT},
-        storage_subpath="graphs/judge",
-    )
-
-Design:
-- ``isolation_level="isolated"`` — the judge has its own memory, not
-  polluting the worker's shared memory namespace.
-- ``conversation_mode="continuous"`` — the judge's conversation carries
-  across timer ticks. The conversation IS the judge's memory. It tracks
-  trends by referring to its own prior messages ("Last check I saw 47
-  steps; now 52; 5 new steps, 3 RETRY").
-- No shared memory keys. No external state files.
-"""
-
-from __future__ import annotations
-
-from framework.graph import Constraint, Goal, NodeSpec, SuccessCriterion
-from framework.graph.edge import AsyncEntryPointSpec, GraphSpec
-
-# ---------------------------------------------------------------------------
-# Goal
-# ---------------------------------------------------------------------------
-
-judge_goal = Goal(
-    id="worker-health-monitor",
-    name="Worker Health Monitor",
-    description=(
-        "Periodically assess the health of the worker agent by reading its "
-        "execution logs. Detect degradation patterns (excessive retries, "
-        "stalls, doom loops) and emit structured EscalationTickets when the "
-        "worker needs attention."
-    ),
-    success_criteria=[
-        SuccessCriterion(
-            id="accurate-detection",
-            description="Only escalates genuine degradation, not normal retry cycles",
-            metric="false_positive_rate",
-            target="low",
-            weight=0.5,
-        ),
-        SuccessCriterion(
-            id="timely-detection",
-            description="Detects genuine stalls within 2 timer ticks (≤4 minutes)",
-            metric="detection_latency_minutes",
-            target="<=4",
-            weight=0.5,
-        ),
-    ],
-    constraints=[
-        Constraint(
-            id="conservative-escalation",
-            description=(
-                "Do not escalate on a single bad verdict or a brief stall. "
-                "Require clear patterns (10+ consecutive bad verdicts or 4+ minute stall) "
-                "before creating a ticket."
-            ),
-            constraint_type="hard",
-            category="quality",
-        ),
-        Constraint(
-            id="complete-ticket",
-            description=(
-                "Every EscalationTicket must have all required fields filled. "
-                "Do not emit partial or placeholder tickets."
-            ),
-            constraint_type="hard",
-            category="correctness",
-        ),
-    ],
-)
-
-# ---------------------------------------------------------------------------
-# Node
-# ---------------------------------------------------------------------------
-
-judge_node = NodeSpec(
-    id="judge",
-    name="Worker Health Judge",
-    description=(
-        "Autonomous health monitor for worker agents. Reads execution logs "
-        "on each timer tick, compares to prior observations (via conversation "
-        "history), and emits a structured EscalationTicket when a genuine "
-        "degradation pattern is detected."
-    ),
-    node_type="event_loop",
-    client_facing=False,  # Autonomous monitor, not interactive
-    max_node_visits=0,  # Unbounded — runs on every timer tick
-    input_keys=[],
-    output_keys=["health_verdict"],
-    nullable_output_keys=["health_verdict"],
-    success_criteria=(
-        "A clear health verdict is produced each check: either 'healthy' with "
-        "a brief observation, or a complete EscalationTicket is emitted via "
-        "emit_escalation_ticket and health_verdict describes the issue."
-    ),
-    tools=[
-        "get_worker_health_summary",
-        "emit_escalation_ticket",
-    ],
-    system_prompt="""\
-You are the Worker Health Judge. You run every 2 minutes alongside a worker \
-agent to monitor its execution health.
-
-# Your Role
-
-You observe the worker's iteration patterns over time and escalate only when \
-you see genuine degradation — not normal retry cycles. Your conversation history \
-IS your memory. On each check, refer to your previous observations to track trends.
-
-# Check Procedure
-
-On each timer tick (every 2 minutes):
-
-## Step 1: Read health snapshot
-Call get_worker_health_summary() with no arguments to auto-discover the active \
-session. This returns:
-- worker_agent_id: the worker's agent name — use this for ticket identity fields
-- worker_graph_id: the worker's primary graph ID — use this for ticket identity fields
-- session_id: the session being monitored — use this for worker_session_id in tickets
-- total_steps: how many log steps have been recorded
-- recent_verdicts: list of recent ACCEPT/RETRY/CONTINUE verdicts
-- steps_since_last_accept: consecutive non-ACCEPT steps
-- stall_minutes: wall-clock since last step (null if active)
-- evidence_snippet: recent LLM output
-
-## Step 2: Compare to prior check
-Look at your conversation history. What was total_steps last time?
-- If total_steps is UNCHANGED from prior check AND prior check was also unchanged:
-  → STALL confirmed (worker has produced no new iterations in 4+ minutes).
-  → Escalate with severity="high" or "critical" depending on stall duration.
-- If total_steps increased: worker is making progress. Examine verdicts.
-
-## Step 3: Analyze verdict pattern
-- Healthy: Mix of ACCEPT and RETRY, steps_since_last_accept < 5. No action.
-- Warning: steps_since_last_accept is 5-9. Note it, no escalation yet.
-- Degraded: steps_since_last_accept >= 10. Examine evidence_snippet.
-  - If evidence shows the agent is making real progress (complex reasoning,
-    exploring solutions, productive tool use): may be a hard problem. Note it.
-  - If evidence shows a loop (same error, same tool call, no new information):
-    → Escalate with severity="medium" or "high".
-- Critical: steps_since_last_accept >= 20, OR stall_minutes >= 4.
-  → Escalate with severity="critical".
-
-## Step 4: Decide
-
-### If healthy:
-set_output("health_verdict", "healthy: <brief observation>")
-Done.
-
-### If escalating:
-Build an EscalationTicket JSON string with ALL required fields:
-{
-  "worker_agent_id": "<worker_agent_id from get_worker_health_summary>",
-  "worker_session_id": "<session_id from get_worker_health_summary>",
-  "worker_node_id": "<worker_graph_id from get_worker_health_summary>",
-  "worker_graph_id": "<worker_graph_id from get_worker_health_summary>",
-  "severity": "<low|medium|high|critical>",
-  "cause": "<what you observed — concrete, specific>",
-  "judge_reasoning": "<why you decided to escalate, not just dismiss>",
-  "suggested_action": "<what you recommend: restart, human review, etc.>",
-  "recent_verdicts": [<list from get_worker_health_summary>],
-  "total_steps_checked": <int>,
-  "steps_since_last_accept": <int>,
-  "stall_minutes": <float or null>,
-  "evidence_snippet": "<from get_worker_health_summary>"
-}
-
-Call: emit_escalation_ticket(ticket_json=<the JSON string above>)
-Then: set_output("health_verdict", "escalated: <one-line summary>")
-
-# Severity Guide
-
-- low: Mild concern, worth noting. 5-9 consecutive bad verdicts.
-- medium: Clear degradation pattern. 10-15 bad verdicts or brief stall (1-2 min).
-- high: Serious issue. 15+ bad verdicts or stall 2-4 minutes or clear doom loop.
-- critical: Worker is definitively stuck. 20+ bad verdicts or stall > 4 minutes.
-
-# Conservative Bias
-
-You MUST resist the urge to escalate prematurely. Worker agents naturally retry.
-A node may legitimately need 5-8 retries before succeeding. Do not escalate unless:
-1. The pattern is clear and sustained across your observation window, AND
-2. The evidence shows no genuine progress
-
-One missed escalation is less costly than two false alarms. The Queen will filter \
-further. But do not be passive — genuine stalls and doom loops must be caught.
-
-# Rules
-- Never escalate on the FIRST check unless stall_minutes > 4
-- Always call get_worker_health_summary FIRST before deciding anything
-- All ticket fields are REQUIRED — do not submit partial tickets
-- After any emit_escalation_ticket call, always set_output to complete the check
-""",
-)
-
-# ---------------------------------------------------------------------------
-# Entry Point
-# ---------------------------------------------------------------------------
-
-HEALTH_JUDGE_ENTRY_POINT = AsyncEntryPointSpec(
-    id="health_check",
-    name="Worker Health Check",
-    entry_node="judge",
-    trigger_type="timer",
-    trigger_config={
-        "interval_minutes": 2,
-        "run_immediately": True,  # Fire immediately to establish a baseline
-    },
-    isolation_level="isolated",  # Own memory namespace, not polluting worker's
-)
-
-# ---------------------------------------------------------------------------
-# Graph
-# ---------------------------------------------------------------------------
-
-judge_graph = GraphSpec(
-    id="judge-graph",
-    goal_id=judge_goal.id,
-    version="1.0.0",
-    entry_node="judge",
-    entry_points={"health_check": "judge"},
-    terminal_nodes=["judge"],  # Judge node can terminate after each check
-    pause_nodes=[],
-    nodes=[judge_node],
-    edges=[],
-    conversation_mode="continuous",  # Conversation persists across timer ticks
-    async_entry_points=[HEALTH_JUDGE_ENTRY_POINT],
-    loop_config={
-        "max_iterations": 10,  # One check shouldn't take many turns
-        "max_tool_calls_per_turn": 3,  # get_summary + optionally emit_ticket
-        "max_history_tokens": 16000,  # Compact — judge only needs recent context
-    },
-)

core/framework/observability/README.md

@@ -83,18 +83,18 @@ configure_logging(level="INFO", format="auto")
 - Compact single-line format (easy to stream/parse)
 - All trace context fields included automatically
 
-### Human-Readable Format (Development)
+### Human-Readable Format (Development / Terminal)
 
 ```
-[INFO    ] [trace:12345678 | exec:a1b2c3d4 | agent:sales-agent] Starting agent execution
-[INFO    ] [trace:12345678 | exec:a1b2c3d4 | agent:sales-agent] Processing input data [node_id:input-processor]
-[INFO    ] [trace:12345678 | exec:a1b2c3d4 | agent:sales-agent] LLM call completed [latency_ms:1250] [tokens_used:450]
+[INFO    ] [agent:sales-agent] Starting agent execution
+[INFO    ] [agent:sales-agent] Processing input data [node_id:input-processor]
+[INFO    ] [agent:sales-agent] LLM call completed [latency_ms:1250] [tokens_used:450]
 ```
 
 **Features:**
 - Color-coded log levels
-- Shortened IDs for readability (first 8 chars)
-- Context prefix shows trace correlation
+- Terminal output omits trace_id and execution_id for readability
+- For full traceability (e.g. debugging), use `ENV=production` to get JSON file logs with trace_id and execution_id
 
 ## Trace Context Fields
 

core/framework/observability/logging.py

@@ -4,8 +4,9 @@ Structured logging with automatic trace context propagation.
 Key Features:
 - Zero developer friction: Standard logger.info() calls get automatic context
 - ContextVar-based propagation: Thread-safe and async-safe
-- Dual output modes: JSON for production, human-readable for development
-- Correlation IDs: trace_id follows entire request flow automatically
+- Dual output modes: JSON for production (full trace_id/execution_id), human-readable for terminal
+- Terminal omits trace_id/execution_id for readability
+- Use ENV=production for file logs with full traceability
 
 Architecture:
     Runtime.start_run() → Generates trace_id, sets context once
@@ -101,10 +102,11 @@ class StructuredFormatter(logging.Formatter):
 
 class HumanReadableFormatter(logging.Formatter):
     """
-    Human-readable formatter for development.
+    Human-readable formatter for development (terminal output).
 
-    Provides colorized logs with trace context for local debugging.
-    Includes trace_id prefix for correlation - AUTOMATIC!
+    Provides colorized logs for local debugging. Omits trace_id and execution_id
+    from the terminal for readability; use ENV=production (JSON file logs) when
+    traceability is needed.
     """
 
     COLORS = {
@@ -118,18 +120,11 @@ class HumanReadableFormatter(logging.Formatter):
 
     def format(self, record: logging.LogRecord) -> str:
         """Format log record as human-readable string."""
-        # Get trace context - AUTOMATIC!
+        # Get trace context; omit trace_id and execution_id in terminal for readability
         context = trace_context.get() or {}
-        trace_id = context.get("trace_id", "")
-        execution_id = context.get("execution_id", "")
         agent_id = context.get("agent_id", "")
 
-        # Build context prefix
         prefix_parts = []
-        if trace_id:
-            prefix_parts.append(f"trace:{trace_id[:8]}")
-        if execution_id:
-            prefix_parts.append(f"exec:{execution_id[-8:]}")
         if agent_id:
             prefix_parts.append(f"agent:{agent_id}")
 
@@ -148,8 +143,9 @@ class HumanReadableFormatter(logging.Formatter):
         if record_event is not None:
             event = f" [{record_event}]"
 
-        # Format message: [LEVEL] [trace context] message
-        return f"{color}[{level}]{reset} {context_prefix}{record.getMessage()}{event}"
+        timestamp = self.formatTime(record, "%Y-%m-%d %H:%M:%S")
+        # Format message: TIMESTAMP [LEVEL] [trace context] message
+        return f"{timestamp} {color}[{level}]{reset} {context_prefix}{record.getMessage()}{event}"
 
 
 def configure_logging(
@@ -210,6 +206,10 @@ def configure_logging(
     root_logger.addHandler(handler)
     root_logger.setLevel(level.upper())
 
+    # Suppress noisy LiteLLM INFO logs (model/provider line + Provider List URL
+    # printed on every single completion call).  Warnings and errors still show.
+    logging.getLogger("LiteLLM").setLevel(logging.WARNING)
+
     # When in JSON mode, configure known third-party loggers to use JSON formatter
     # This ensures libraries like LiteLLM, httpcore also output clean JSON
     if format == "json":

core/framework/runner/cli.py

@@ -243,6 +243,8 @@ def register_commands(subparsers: argparse._SubParsersAction) -> None:
         action="store_true",
         help="Open dashboard in browser after server starts",
     )
+    serve_parser.add_argument("--verbose", "-v", action="store_true", help="Enable INFO log level")
+    serve_parser.add_argument("--debug", action="store_true", help="Enable DEBUG log level")
     serve_parser.set_defaults(func=cmd_serve)
 
     # open command (serve + auto-open browser)
@@ -280,6 +282,8 @@ def register_commands(subparsers: argparse._SubParsersAction) -> None:
         default=None,
         help="LLM model for preloaded agents",
     )
+    open_parser.add_argument("--verbose", "-v", action="store_true", help="Enable INFO log level")
+    open_parser.add_argument("--debug", action="store_true", help="Enable DEBUG log level")
     open_parser.set_defaults(func=cmd_open)
 
 
@@ -375,18 +379,18 @@ def _prompt_before_start(agent_path: str, runner, model: str | None = None):
 
 def cmd_run(args: argparse.Namespace) -> int:
     """Run an exported agent."""
-    import logging
 
     from framework.credentials.models import CredentialError
+    from framework.observability import configure_logging
     from framework.runner import AgentRunner
 
     # Set logging level (quiet by default for cleaner output)
     if args.quiet:
-        logging.basicConfig(level=logging.ERROR, format="%(message)s")
+        configure_logging(level="ERROR")
     elif getattr(args, "verbose", False):
-        logging.basicConfig(level=logging.INFO, format="%(message)s")
+        configure_logging(level="INFO")
     else:
-        logging.basicConfig(level=logging.WARNING, format="%(message)s")
+        configure_logging(level="WARNING")
 
     # Load input context
     context = {}
@@ -742,6 +746,17 @@ def cmd_dispatch(args: argparse.Namespace) -> int:
     if args.agents:
         # Use specific agents
         for agent_name in args.agents:
+            # Guard against full paths: if the name contains path separators
+            # (e.g. "exports/my_agent"), it will be doubled with agents_dir
+            agent_name_path = Path(agent_name)
+            if len(agent_name_path.parts) > 1:
+                print(
+                    f"Error: --agents expects agent names, not paths. "
+                    f"Use: --agents {agent_name_path.name} "
+                    f"instead of --agents {agent_name}",
+                    file=sys.stderr,
+                )
+                return 1
             agent_path = agents_dir / agent_name
             if not _is_valid_agent_dir(agent_path):
                 print(f"Agent not found: {agent_path}", file=sys.stderr)
@@ -907,16 +922,12 @@ def _format_natural_language_to_json(
 
 def cmd_shell(args: argparse.Namespace) -> int:
     """Start an interactive agent session."""
-    import logging
 
     from framework.credentials.models import CredentialError
+    from framework.observability import configure_logging
     from framework.runner import AgentRunner
 
-    # Configure logging to show runtime visibility
-    logging.basicConfig(
-        level=logging.INFO,
-        format="%(message)s",  # Simple format for clean output
-    )
+    configure_logging(level="INFO")
 
     agents_dir = Path(args.agents_dir)
 
@@ -1614,18 +1625,18 @@ def _build_frontend() -> bool:
 
 def cmd_serve(args: argparse.Namespace) -> int:
     """Start the HTTP API server."""
-    import logging
 
     from aiohttp import web
 
     _build_frontend()
 
+    from framework.observability import configure_logging
     from framework.server.app import create_app
 
-    logging.basicConfig(
-        level=logging.INFO,
-        format="%(asctime)s [%(levelname)s] %(name)s: %(message)s",
-    )
+    if getattr(args, "debug", False):
+        configure_logging(level="DEBUG")
+    else:
+        configure_logging(level="INFO")
 
     model = getattr(args, "model", None)
     app = create_app(model=model)

core/framework/runner/runner.py

@@ -9,14 +9,13 @@ from datetime import UTC
 from pathlib import Path
 from typing import TYPE_CHECKING, Any
 
-from framework.config import get_hive_config, get_preferred_model
+from framework.config import get_hive_config, get_max_context_tokens, get_preferred_model
 from framework.credentials.validation import (
     ensure_credential_key_env as _ensure_credential_key_env,
 )
 from framework.graph import Goal
 from framework.graph.edge import (
     DEFAULT_MAX_TOKENS,
-    AsyncEntryPointSpec,
     EdgeCondition,
     EdgeSpec,
     GraphSpec,
@@ -29,6 +28,7 @@ from framework.runner.tool_registry import ToolRegistry
 from framework.runtime.agent_runtime import AgentRuntime, AgentRuntimeConfig, create_agent_runtime
 from framework.runtime.execution_stream import EntryPointSpec
 from framework.runtime.runtime_log_store import RuntimeLogStore
+from framework.tools.flowchart_utils import generate_fallback_flowchart
 
 if TYPE_CHECKING:
     from framework.runner.protocol import AgentMessage, CapabilityResponse
@@ -570,9 +570,6 @@ class AgentInfo:
     constraints: list[dict]
     required_tools: list[str]
     has_tools_module: bool
-    # Multi-entry-point support
-    async_entry_points: list[dict] = field(default_factory=list)
-    is_multi_entry_point: bool = False
 
 
 @dataclass
@@ -630,22 +627,6 @@ def load_agent_export(data: str | dict) -> tuple[GraphSpec, Goal]:
         )
         edges.append(edge)
 
-    # Build AsyncEntryPointSpec objects for multi-entry-point support
-    async_entry_points = []
-    for aep_data in graph_data.get("async_entry_points", []):
-        async_entry_points.append(
-            AsyncEntryPointSpec(
-                id=aep_data["id"],
-                name=aep_data.get("name", aep_data["id"]),
-                entry_node=aep_data["entry_node"],
-                trigger_type=aep_data.get("trigger_type", "manual"),
-                trigger_config=aep_data.get("trigger_config", {}),
-                isolation_level=aep_data.get("isolation_level", "shared"),
-                priority=aep_data.get("priority", 0),
-                max_concurrent=aep_data.get("max_concurrent", 10),
-            )
-        )
-
     # Build GraphSpec
     graph = GraphSpec(
         id=graph_data.get("id", "agent-graph"),
@@ -653,7 +634,6 @@ def load_agent_export(data: str | dict) -> tuple[GraphSpec, Goal]:
         version=graph_data.get("version", "1.0.0"),
         entry_node=graph_data.get("entry_node", ""),
         entry_points=graph_data.get("entry_points", {}),  # Support pause/resume architecture
-        async_entry_points=async_entry_points,  # Support multi-entry-point agents
         terminal_nodes=graph_data.get("terminal_nodes", []),
         pause_nodes=graph_data.get("pause_nodes", []),  # Support pause/resume architecture
         nodes=nodes,
@@ -805,8 +785,6 @@ class AgentRunner:
 
         # AgentRuntime — unified execution path for all agents
         self._agent_runtime: AgentRuntime | None = None
-        self._uses_async_entry_points = self.graph.has_async_entry_points()
-
         # Pre-load validation: structural checks + credentials.
         # Fails fast with actionable guidance — no MCP noise on screen.
         run_preload_validation(
@@ -926,10 +904,32 @@ class AgentRunner:
 
             if agent_config and hasattr(agent_config, "max_tokens"):
                 max_tokens = agent_config.max_tokens
+                logger.info(
+                    "Agent default_config overrides max_tokens: %d "
+                    "(configuration.json value ignored)",
+                    max_tokens,
+                )
             else:
                 hive_config = get_hive_config()
                 max_tokens = hive_config.get("llm", {}).get("max_tokens", DEFAULT_MAX_TOKENS)
 
+            # Resolve max_context_tokens with priority:
+            #   1. agent loop_config["max_context_tokens"] (explicit, wins silently)
+            #   2. agent default_config.max_context_tokens (logged)
+            #   3. configuration.json llm.max_context_tokens
+            #   4. hardcoded default (32_000)
+            agent_loop_config: dict = dict(getattr(agent_module, "loop_config", {}))
+            if "max_context_tokens" not in agent_loop_config:
+                if agent_config and hasattr(agent_config, "max_context_tokens"):
+                    agent_loop_config["max_context_tokens"] = agent_config.max_context_tokens
+                    logger.info(
+                        "Agent default_config overrides max_context_tokens: %d"
+                        " (configuration.json value ignored)",
+                        agent_config.max_context_tokens,
+                    )
+                else:
+                    agent_loop_config["max_context_tokens"] = get_max_context_tokens()
+
             # Read intro_message from agent metadata (shown on TUI load)
             agent_metadata = getattr(agent_module, "metadata", None)
             intro_message = ""
@@ -943,13 +943,12 @@ class AgentRunner:
                 "version": "1.0.0",
                 "entry_node": getattr(agent_module, "entry_node", nodes[0].id),
                 "entry_points": getattr(agent_module, "entry_points", {}),
-                "async_entry_points": getattr(agent_module, "async_entry_points", []),
                 "terminal_nodes": getattr(agent_module, "terminal_nodes", []),
                 "pause_nodes": getattr(agent_module, "pause_nodes", []),
                 "nodes": nodes,
                 "edges": edges,
                 "max_tokens": max_tokens,
-                "loop_config": getattr(agent_module, "loop_config", {}),
+                "loop_config": agent_loop_config,
             }
             # Only pass optional fields if explicitly defined by the agent module
             conversation_mode = getattr(agent_module, "conversation_mode", None)
@@ -961,6 +960,12 @@ class AgentRunner:
 
             graph = GraphSpec(**graph_kwargs)
 
+            # Generate flowchart.json if missing (for template/legacy agents)
+            generate_fallback_flowchart(graph, goal, agent_path)
+            # Read skill configuration from agent module
+            agent_default_skills = getattr(agent_module, "default_skills", None)
+            agent_skills = getattr(agent_module, "skills", None)
+
             # Read runtime config (webhook settings, etc.) if defined
             agent_runtime_config = getattr(agent_module, "runtime_config", None)
 
@@ -972,7 +977,7 @@ class AgentRunner:
             configure_fn = getattr(agent_module, "configure_for_account", None)
             list_accts_fn = getattr(agent_module, "list_connected_accounts", None)
 
-            return cls(
+            runner = cls(
                 agent_path=agent_path,
                 graph=graph,
                 goal=goal,
@@ -988,6 +993,10 @@ class AgentRunner:
                 list_accounts=list_accts_fn,
                 credential_store=credential_store,
             )
+            # Stash skill config for use in _setup()
+            runner._agent_default_skills = agent_default_skills
+            runner._agent_skills = agent_skills
+            return runner
 
         # Fallback: load from agent.json (legacy JSON-based agents)
         agent_json_path = agent_path / "agent.json"
@@ -1005,7 +1014,10 @@ class AgentRunner:
         except json.JSONDecodeError as exc:
             raise ValueError(f"Invalid JSON in agent export file: {agent_json_path}") from exc
 
-        return cls(
+        # Generate flowchart.json if missing (for legacy JSON-based agents)
+        generate_fallback_flowchart(graph, goal, agent_path)
+
+        runner = cls(
             agent_path=agent_path,
             graph=graph,
             goal=goal,
@@ -1016,6 +1028,9 @@ class AgentRunner:
             skip_credential_validation=skip_credential_validation or False,
             credential_store=credential_store,
         )
+        runner._agent_default_skills = None
+        runner._agent_skills = None
+        return runner
 
     def register_tool(
         self,
@@ -1325,6 +1340,19 @@ class AgentRunner:
         except Exception:
             pass  # Best-effort — agent works without account info
 
+        # Skill configuration — the runtime handles discovery, loading, and
+        # prompt rasterization.  The runner just builds the config.
+        from framework.skills.config import SkillsConfig
+        from framework.skills.manager import SkillsManagerConfig
+
+        skills_manager_config = SkillsManagerConfig(
+            skills_config=SkillsConfig.from_agent_vars(
+                default_skills=getattr(self, "_agent_default_skills", None),
+                skills=getattr(self, "_agent_skills", None),
+            ),
+            project_root=self.agent_path,
+        )
+
         self._setup_agent_runtime(
             tools,
             tool_executor,
@@ -1332,6 +1360,7 @@ class AgentRunner:
             accounts_data=accounts_data,
             tool_provider_map=tool_provider_map,
             event_bus=event_bus,
+            skills_manager_config=skills_manager_config,
         )
 
     def _get_api_key_env_var(self, model: str) -> str | None:
@@ -1352,6 +1381,8 @@ class AgentRunner:
             return "MISTRAL_API_KEY"
         elif model_lower.startswith("groq/"):
             return "GROQ_API_KEY"
+        elif model_lower.startswith("openrouter/"):
+            return "OPENROUTER_API_KEY"
         elif self._is_local_model(model_lower):
             return None  # Local models don't need an API key
         elif model_lower.startswith("azure/"):
@@ -1366,6 +1397,8 @@ class AgentRunner:
             return "MINIMAX_API_KEY"
         elif model_lower.startswith("kimi/"):
             return "KIMI_API_KEY"
+        elif model_lower.startswith("hive/"):
+            return "HIVE_API_KEY"
         else:
             # Default: assume OpenAI-compatible
             return "OPENAI_API_KEY"
@@ -1388,6 +1421,8 @@ class AgentRunner:
             cred_id = "minimax"
         elif model_lower.startswith("kimi/"):
             cred_id = "kimi"
+        elif model_lower.startswith("hive/"):
+            cred_id = "hive"
         # Add more mappings as providers are added to LLM_CREDENTIALS
 
         if cred_id is None:
@@ -1427,23 +1462,10 @@ class AgentRunner:
         accounts_data: list[dict] | None = None,
         tool_provider_map: dict[str, str] | None = None,
         event_bus=None,
+        skills_manager_config=None,
     ) -> None:
         """Set up multi-entry-point execution using AgentRuntime."""
-        # Convert AsyncEntryPointSpec to EntryPointSpec for AgentRuntime
         entry_points = []
-        for async_ep in self.graph.async_entry_points:
-            ep = EntryPointSpec(
-                id=async_ep.id,
-                name=async_ep.name,
-                entry_node=async_ep.entry_node,
-                trigger_type=async_ep.trigger_type,
-                trigger_config=async_ep.trigger_config,
-                isolation_level=async_ep.isolation_level,
-                priority=async_ep.priority,
-                max_concurrent=async_ep.max_concurrent,
-                max_resurrections=async_ep.max_resurrections,
-            )
-            entry_points.append(ep)
 
         # Always create a primary entry point for the graph's entry node.
         # For multi-entry-point agents this ensures the primary path (e.g.
@@ -1500,26 +1522,37 @@ class AgentRunner:
             accounts_data=accounts_data,
             tool_provider_map=tool_provider_map,
             event_bus=event_bus,
+            skills_manager_config=skills_manager_config,
         )
 
         # Pass intro_message through for TUI display
         self._agent_runtime.intro_message = self.intro_message
 
+    # ------------------------------------------------------------------
+    # Execution modes
+    #
+    # run()              – One-shot, blocking execution for worker agents
+    #                      (headless CLI via ``hive run``). Validates, runs
+    #                      the graph to completion, and returns the result.
+    #
+    # start() / trigger() – Long-lived runtime for the frontend (queen).
+    #                      start() boots the runtime; trigger() sends
+    #                      non-blocking execution requests. Used by the
+    #                      server session manager and API routes.
+    # ------------------------------------------------------------------
+
     async def run(
         self,
         input_data: dict | None = None,
         session_state: dict | None = None,
         entry_point_id: str | None = None,
     ) -> ExecutionResult:
-        """
-        Execute the agent with given input data.
+        """One-shot execution for worker agents (headless CLI).
 
-        Validates credentials before execution. If any required credentials
-        are missing, returns an error result with instructions on how to
-        provide them.
+        Validates credentials, runs the graph to completion, and returns
+        the result. Used by ``hive run`` and programmatic callers.
 
-        For single-entry-point agents, this is the standard execution path.
-        For multi-entry-point agents, you can optionally specify which entry point to use.
+        For the frontend (queen), use start() + trigger() instead.
 
         Args:
             input_data: Input data for the agent (e.g., {"lead_id": "123"})
@@ -1645,7 +1678,12 @@ class AgentRunner:
     # === Runtime API ===
 
     async def start(self) -> None:
-        """Start the agent runtime."""
+        """Boot the agent runtime for the frontend (queen).
+
+        Pair with trigger() to send execution requests. Used by the
+        server session manager. For headless worker agents, use run()
+        instead.
+        """
         if self._agent_runtime is None:
             self._setup()
 
@@ -1662,10 +1700,10 @@ class AgentRunner:
         input_data: dict[str, Any],
         correlation_id: str | None = None,
     ) -> str:
-        """
-        Trigger execution at a specific entry point (non-blocking).
+        """Send a non-blocking execution request to a running runtime.
 
-        Returns execution ID for tracking.
+        Used by the server API routes after start(). For headless
+        worker agents, use run() instead.
 
         Args:
             entry_point_id: Which entry point to trigger
@@ -1750,19 +1788,6 @@ class AgentRunner:
             for edge in self.graph.edges
         ]
 
-        # Build async entry points info
-        async_entry_points_info = [
-            {
-                "id": ep.id,
-                "name": ep.name,
-                "entry_node": ep.entry_node,
-                "trigger_type": ep.trigger_type,
-                "isolation_level": ep.isolation_level,
-                "max_concurrent": ep.max_concurrent,
-            }
-            for ep in self.graph.async_entry_points
-        ]
-
         return AgentInfo(
             name=self.graph.id,
             description=self.graph.description,
@@ -1789,8 +1814,6 @@ class AgentRunner:
             ],
             required_tools=sorted(required_tools),
             has_tools_module=(self.agent_path / "tools.py").exists(),
-            async_entry_points=async_entry_points_info,
-            is_multi_entry_point=self._uses_async_entry_points,
         )
 
     def validate(self) -> ValidationResult:
@@ -2105,18 +2128,6 @@ Respond with JSON only:
                 trigger_type="manual",
                 isolation_level="shared",
             )
-        for aep in runner.graph.async_entry_points:
-            entry_points[aep.id] = EntryPointSpec(
-                id=aep.id,
-                name=aep.name,
-                entry_node=aep.entry_node,
-                trigger_type=aep.trigger_type,
-                trigger_config=aep.trigger_config,
-                isolation_level=aep.isolation_level,
-                priority=aep.priority,
-                max_concurrent=aep.max_concurrent,
-            )
-
         await runtime.add_graph(
             graph_id=gid,
             graph=runner.graph,

core/framework/runner/tool_registry.py

@@ -455,11 +455,23 @@ class ToolRegistry:
 
         for server_config in server_list:
             server_config = self._resolve_mcp_server_config(server_config, base_dir)
-            try:
-                self.register_mcp_server(server_config)
-            except Exception as e:
-                name = server_config.get("name", "unknown")
-                logger.warning(f"Failed to register MCP server '{name}': {e}")
+            for _attempt in range(2):
+                try:
+                    self.register_mcp_server(server_config)
+                    break
+                except Exception as e:
+                    name = server_config.get("name", "unknown")
+                    if _attempt == 0:
+                        logger.warning(
+                            "MCP server '%s' failed to register, retrying in 2s: %s",
+                            name,
+                            e,
+                        )
+                        import time
+
+                        time.sleep(2)
+                    else:
+                        logger.warning("MCP server '%s' failed after retry: %s", name, e)
 
         # Snapshot credential files and ADEN_API_KEY so we can detect mid-session changes
         self._mcp_cred_snapshot = self._snapshot_credentials()

core/framework/runtime/EVENT_TYPES.md

@@ -454,11 +454,11 @@ An agent has requested handoff to the Hive Coder (via the `escalate` synthetic t
 
 ## Worker Health Monitoring
 
-These events form the **judge → queen → operator** escalation pipeline.
+These events form the **queen → operator** escalation pipeline.
 
 ### `worker_escalation_ticket`
 
-The Worker Health Judge has detected a degradation pattern and is escalating to the Queen.
+A worker degradation pattern has been detected and is being escalated to the Queen.
 
 | Data Field | Type   | Description                          |
 | ---------- | ------ | ------------------------------------ |

core/framework/runtime/agent_runtime.py

@@ -8,6 +8,7 @@ while preserving the goal-driven approach.
 import asyncio
 import logging
 import time
+import uuid
 from collections.abc import Callable
 from dataclasses import dataclass, field
 from datetime import datetime
@@ -28,6 +29,7 @@ if TYPE_CHECKING:
     from framework.graph.edge import GraphSpec
     from framework.graph.goal import Goal
     from framework.llm.provider import LLMProvider, Tool
+    from framework.skills.manager import SkillsManagerConfig
 
 logger = logging.getLogger(__name__)
 
@@ -131,6 +133,10 @@ class AgentRuntime:
         accounts_data: list[dict] | None = None,
         tool_provider_map: dict[str, str] | None = None,
         event_bus: "EventBus | None" = None,
+        skills_manager_config: "SkillsManagerConfig | None" = None,
+        # Deprecated — pass skills_manager_config instead.
+        skills_catalog_prompt: str = "",
+        protocols_prompt: str = "",
     ):
         """
         Initialize agent runtime.
@@ -152,7 +158,13 @@ class AgentRuntime:
             event_bus: Optional external EventBus. If provided, the runtime shares
                 this bus instead of creating its own. Used by SessionManager to
                 share a single bus between queen, worker, and judge.
+            skills_manager_config: Skill configuration — the runtime owns
+                discovery, loading, and prompt renderation internally.
+            skills_catalog_prompt: Deprecated. Pre-rendered skills catalog.
+            protocols_prompt: Deprecated. Pre-rendered operational protocols.
         """
+        from framework.skills.manager import SkillsManager
+
         self.graph = graph
         self.goal = goal
         self._config = config or AgentRuntimeConfig()
@@ -160,6 +172,29 @@ class AgentRuntime:
         self._checkpoint_config = checkpoint_config
         self.accounts_prompt = accounts_prompt
 
+        # --- Skill lifecycle: runtime owns the SkillsManager ---
+        if skills_manager_config is not None:
+            # New path: config-driven, runtime handles loading
+            self._skills_manager = SkillsManager(skills_manager_config)
+            self._skills_manager.load()
+        elif skills_catalog_prompt or protocols_prompt:
+            # Legacy path: caller passed pre-rendered strings
+            import warnings
+
+            warnings.warn(
+                "Passing pre-rendered skills_catalog_prompt/protocols_prompt "
+                "is deprecated. Pass skills_manager_config instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
+            self._skills_manager = SkillsManager.from_precomputed(
+                skills_catalog_prompt, protocols_prompt
+            )
+        else:
+            # Bare constructor: auto-load defaults
+            self._skills_manager = SkillsManager()
+            self._skills_manager.load()
+
         # Primary graph identity
         self._graph_id: str = graph_id or "primary"
 
@@ -215,6 +250,18 @@ class AgentRuntime:
         # Optional greeting shown to user on TUI load (set by AgentRunner)
         self.intro_message: str = ""
 
+    # ------------------------------------------------------------------
+    # Skill prompt accessors (read by ExecutionStream constructors)
+    # ------------------------------------------------------------------
+
+    @property
+    def skills_catalog_prompt(self) -> str:
+        return self._skills_manager.skills_catalog_prompt
+
+    @property
+    def protocols_prompt(self) -> str:
+        return self._skills_manager.protocols_prompt
+
     def register_entry_point(self, spec: EntryPointSpec) -> None:
         """
         Register a named entry point for the agent.
@@ -292,6 +339,8 @@ class AgentRuntime:
                     accounts_prompt=self._accounts_prompt,
                     accounts_data=self._accounts_data,
                     tool_provider_map=self._tool_provider_map,
+                    skills_catalog_prompt=self.skills_catalog_prompt,
+                    protocols_prompt=self.protocols_prompt,
                 )
                 await stream.start()
                 self._streams[ep_id] = stream
@@ -392,18 +441,24 @@ class AgentRuntime:
 
                 tc = spec.trigger_config
                 cron_expr = tc.get("cron")
-                interval = tc.get("interval_minutes")
+                _raw_interval = tc.get("interval_minutes")
+                interval = float(_raw_interval) if _raw_interval is not None else None
                 run_immediately = tc.get("run_immediately", False)
 
                 if cron_expr:
                     # Cron expression mode — takes priority over interval_minutes
                     try:
                         from croniter import croniter
+                    except ImportError as e:
+                        raise RuntimeError(
+                            "croniter is required for cron-based entry points. "
+                            "Install it with: uv pip install croniter"
+                        ) from e
 
-                        # Validate the expression upfront
+                    try:
                         if not croniter.is_valid(cron_expr):
                             raise ValueError(f"Invalid cron expression: {cron_expr}")
-                    except (ImportError, ValueError) as e:
+                    except ValueError as e:
                         logger.warning(
                             "Entry point '%s' has invalid cron config: %s",
                             ep_id,
@@ -543,7 +598,7 @@ class AgentRuntime:
                             ep_id,
                             cron_expr,
                             run_immediately,
-                            idle_timeout=tc.get("idle_timeout_seconds", 300),
+                            idle_timeout=float(tc.get("idle_timeout_seconds", 300)),
                         )()
                     )
                     self._timer_tasks.append(task)
@@ -673,7 +728,7 @@ class AgentRuntime:
                             ep_id,
                             interval,
                             run_immediately,
-                            idle_timeout=tc.get("idle_timeout_seconds", 300),
+                            idle_timeout=float(tc.get("idle_timeout_seconds", 300)),
                         )()
                     )
                     self._timer_tasks.append(task)
@@ -822,7 +877,8 @@ class AgentRuntime:
         if stream is None:
             raise ValueError(f"Entry point '{entry_point_id}' not found")
 
-        return await stream.execute(input_data, correlation_id, session_state)
+        run_id = uuid.uuid4().hex[:12]
+        return await stream.execute(input_data, correlation_id, session_state, run_id=run_id)
 
     async def trigger_and_wait(
         self,
@@ -919,6 +975,8 @@ class AgentRuntime:
                 accounts_prompt=self._accounts_prompt,
                 accounts_data=self._accounts_data,
                 tool_provider_map=self._tool_provider_map,
+                skills_catalog_prompt=self.skills_catalog_prompt,
+                protocols_prompt=self.protocols_prompt,
             )
             if self._running:
                 await stream.start()
@@ -997,7 +1055,8 @@ class AgentRuntime:
             if spec.trigger_type != "timer":
                 continue
             tc = spec.trigger_config
-            interval = tc.get("interval_minutes")
+            _raw_interval = tc.get("interval_minutes")
+            interval = float(_raw_interval) if _raw_interval is not None else None
             run_immediately = tc.get("run_immediately", False)
 
             if interval and interval > 0 and self._running:
@@ -1142,7 +1201,7 @@ class AgentRuntime:
                         ep_id,
                         interval,
                         run_immediately,
-                        idle_timeout=tc.get("idle_timeout_seconds", 300),
+                        idle_timeout=float(tc.get("idle_timeout_seconds", 300)),
                     )()
                 )
                 timer_tasks.append(task)
@@ -1359,8 +1418,8 @@ class AgentRuntime:
                 allowed_keys = set(entry_node.input_keys)
 
         # Search primary graph's streams for an active session.
-        # Skip isolated streams (e.g. health judge) — they have their own
-        # session directories and must never be used as a shared session.
+        # Skip isolated streams — they have their own session directories
+        # and must never be used as a shared session.
         all_streams: list[tuple[str, ExecutionStream]] = []
         for _gid, reg in self._graphs.items():
             for ep_id, stream in reg.streams.items():
@@ -1697,6 +1756,10 @@ def create_agent_runtime(
     accounts_data: list[dict] | None = None,
     tool_provider_map: dict[str, str] | None = None,
     event_bus: "EventBus | None" = None,
+    skills_manager_config: "SkillsManagerConfig | None" = None,
+    # Deprecated — pass skills_manager_config instead.
+    skills_catalog_prompt: str = "",
+    protocols_prompt: str = "",
 ) -> AgentRuntime:
     """
     Create and configure an AgentRuntime with entry points.
@@ -1723,6 +1786,10 @@ def create_agent_runtime(
         accounts_data: Raw account data for per-node prompt generation.
         tool_provider_map: Tool name to provider name mapping for account routing.
         event_bus: Optional external EventBus to share with other components.
+        skills_manager_config: Skill configuration — the runtime owns
+            discovery, loading, and prompt renderation internally.
+        skills_catalog_prompt: Deprecated. Pre-rendered skills catalog.
+        protocols_prompt: Deprecated. Pre-rendered operational protocols.
 
     Returns:
         Configured AgentRuntime (not yet started)
@@ -1749,6 +1816,9 @@ def create_agent_runtime(
         accounts_data=accounts_data,
         tool_provider_map=tool_provider_map,
         event_bus=event_bus,
+        skills_manager_config=skills_manager_config,
+        skills_catalog_prompt=skills_catalog_prompt,
+        protocols_prompt=protocols_prompt,
     )
 
     for spec in entry_points:

core/framework/runtime/escalation_ticket.py

@@ -1,4 +1,4 @@
-"""EscalationTicket — structured schema for worker health judge escalations."""
+"""EscalationTicket — structured schema for worker health escalations."""
 
 from __future__ import annotations
 
@@ -10,10 +10,10 @@ from pydantic import BaseModel, Field
 
 
 class EscalationTicket(BaseModel):
-    """Structured escalation report emitted by the Worker Health Judge.
+    """Structured escalation report for worker health monitoring.
 
-    The judge must fill every field before calling emit_escalation_ticket.
-    Pydantic validation rejects partial tickets, preventing impulsive escalation.
+    All fields must be filled before calling emit_escalation_ticket.
+    Pydantic validation rejects partial tickets.
     """
 
     ticket_id: str = Field(default_factory=lambda: str(uuid4()))
@@ -25,7 +25,7 @@ class EscalationTicket(BaseModel):
     worker_node_id: str
     worker_graph_id: str
 
-    # Problem characterization (filled by judge via LLM deliberation)
+    # Problem characterization
     severity: Literal["low", "medium", "high", "critical"]
     cause: str  # Human-readable: "Node has produced 18 RETRY verdicts..."
     judge_reasoning: str  # Judge's own deliberation chain

core/framework/runtime/event_bus.py

@@ -97,6 +97,7 @@ class EventType(StrEnum):
     # Client I/O (client_facing=True nodes only)
     CLIENT_OUTPUT_DELTA = "client_output_delta"
     CLIENT_INPUT_REQUESTED = "client_input_requested"
+    CLIENT_INPUT_RECEIVED = "client_input_received"
 
     # Internal node observability (client_facing=False nodes)
     NODE_INTERNAL_OUTPUT = "node_internal_output"
@@ -104,7 +105,7 @@ class EventType(StrEnum):
     NODE_STALLED = "node_stalled"
     NODE_TOOL_DOOM_LOOP = "node_tool_doom_loop"
 
-    # Judge decisions
+    # Judge decisions (implicit judge in event loop nodes)
     JUDGE_VERDICT = "judge_verdict"
 
     # Output tracking
@@ -126,7 +127,7 @@ class EventType(StrEnum):
     # Escalation (agent requests handoff to queen)
     ESCALATION_REQUESTED = "escalation_requested"
 
-    # Worker health monitoring (judge → queen → operator)
+    # Worker health monitoring
     WORKER_ESCALATION_TICKET = "worker_escalation_ticket"
     QUEEN_INTERVENTION_REQUESTED = "queen_intervention_requested"
 
@@ -137,6 +138,12 @@ class EventType(StrEnum):
     WORKER_LOADED = "worker_loaded"
     CREDENTIALS_REQUIRED = "credentials_required"
 
+    # Draft graph (planning phase — lightweight graph preview)
+    DRAFT_GRAPH_UPDATED = "draft_graph_updated"
+
+    # Flowchart map updated (after reconciliation with runtime graph)
+    FLOWCHART_MAP_UPDATED = "flowchart_map_updated"
+
     # Queen phase changes (building <-> staging <-> running)
     QUEEN_PHASE_CHANGED = "queen_phase_changed"
 
@@ -146,6 +153,14 @@ class EventType(StrEnum):
     # Subagent reports (one-way progress updates from sub-agents)
     SUBAGENT_REPORT = "subagent_report"
 
+    # Trigger lifecycle (queen-level triggers / heartbeats)
+    TRIGGER_AVAILABLE = "trigger_available"
+    TRIGGER_ACTIVATED = "trigger_activated"
+    TRIGGER_DEACTIVATED = "trigger_deactivated"
+    TRIGGER_FIRED = "trigger_fired"
+    TRIGGER_REMOVED = "trigger_removed"
+    TRIGGER_UPDATED = "trigger_updated"
+
 
 @dataclass
 class AgentEvent:
@@ -159,10 +174,11 @@ class AgentEvent:
     timestamp: datetime = field(default_factory=datetime.now)
     correlation_id: str | None = None  # For tracking related events
     graph_id: str | None = None  # Which graph emitted this event (multi-graph sessions)
+    run_id: str | None = None  # Unique ID per trigger() invocation — used for run dividers
 
     def to_dict(self) -> dict:
         """Convert to dictionary for serialization."""
-        return {
+        d = {
             "type": self.type.value,
             "stream_id": self.stream_id,
             "node_id": self.node_id,
@@ -172,6 +188,9 @@ class AgentEvent:
             "correlation_id": self.correlation_id,
             "graph_id": self.graph_id,
         }
+        if self.run_id is not None:
+            d["run_id"] = self.run_id
+        return d
 
 
 # Type for event handlers
@@ -240,6 +259,128 @@ class EventBus:
         self._semaphore = asyncio.Semaphore(max_concurrent_handlers)
         self._subscription_counter = 0
         self._lock = asyncio.Lock()
+        # Per-session persistent event log (always-on, survives restarts)
+        self._session_log: IO[str] | None = None
+        self._session_log_iteration_offset: int = 0
+        # Accumulator for client_output_delta snapshots — flushed on llm_turn_complete.
+        # Key: (stream_id, node_id, execution_id, iteration, inner_turn) → latest AgentEvent
+        self._pending_output_snapshots: dict[tuple, AgentEvent] = {}
+
+    def set_session_log(self, path: Path, *, iteration_offset: int = 0) -> None:
+        """Enable per-session event persistence to a JSONL file.
+
+        Called once when the queen starts so that all events survive server
+        restarts and can be replayed to reconstruct the frontend state.
+
+        ``iteration_offset`` is added to the ``iteration`` field in logged
+        events so that cold-resumed sessions produce monotonically increasing
+        iteration values — preventing frontend message ID collisions between
+        the original run and resumed runs.
+        """
+        if self._session_log is not None:
+            try:
+                self._session_log.close()
+            except Exception:
+                pass
+        path.parent.mkdir(parents=True, exist_ok=True)
+        self._session_log = open(path, "a", encoding="utf-8")  # noqa: SIM115
+        self._session_log_iteration_offset = iteration_offset
+        logger.info("Session event log → %s (iteration_offset=%d)", path, iteration_offset)
+
+    def close_session_log(self) -> None:
+        """Close the per-session event log file."""
+        # Flush any pending output snapshots before closing
+        self._flush_pending_snapshots()
+        if self._session_log is not None:
+            try:
+                self._session_log.close()
+            except Exception:
+                pass
+            self._session_log = None
+
+    # Event types that are high-frequency streaming deltas — accumulated rather
+    # than written individually to the session log.
+    _STREAMING_DELTA_TYPES = frozenset(
+        {
+            EventType.CLIENT_OUTPUT_DELTA,
+            EventType.LLM_TEXT_DELTA,
+            EventType.LLM_REASONING_DELTA,
+        }
+    )
+
+    def _write_session_log_event(self, event: AgentEvent) -> None:
+        """Write an event to the per-session log with streaming coalescing.
+
+        Streaming deltas (client_output_delta, llm_text_delta) are accumulated
+        in memory.  When llm_turn_complete fires, any pending snapshots for that
+        (stream_id, node_id, execution_id) are flushed as single consolidated
+        events before the turn-complete event itself is written.
+
+        Note: iteration offset is already applied in publish() before this is
+        called, so events here already have correct iteration values.
+        """
+        if self._session_log is None:
+            return
+
+        if event.type in self._STREAMING_DELTA_TYPES:
+            # Accumulate — keep only the latest event (which carries the full snapshot)
+            key = (
+                event.stream_id,
+                event.node_id,
+                event.execution_id,
+                event.data.get("iteration"),
+                event.data.get("inner_turn", 0),
+            )
+            self._pending_output_snapshots[key] = event
+            return
+
+        # On turn-complete, flush accumulated snapshots for this stream first
+        if event.type == EventType.LLM_TURN_COMPLETE:
+            self._flush_pending_snapshots(
+                stream_id=event.stream_id,
+                node_id=event.node_id,
+                execution_id=event.execution_id,
+            )
+
+        line = json.dumps(event.to_dict(), default=str)
+        self._session_log.write(line + "\n")
+        self._session_log.flush()
+
+    def _flush_pending_snapshots(
+        self,
+        stream_id: str | None = None,
+        node_id: str | None = None,
+        execution_id: str | None = None,
+    ) -> None:
+        """Flush accumulated streaming snapshots to the session log.
+
+        When called with filters, only matching entries are flushed.
+        When called without filters (e.g. on close), everything is flushed.
+        """
+        if self._session_log is None or not self._pending_output_snapshots:
+            return
+
+        to_flush: list[tuple] = []
+        for key, _evt in self._pending_output_snapshots.items():
+            if stream_id is not None:
+                k_stream, k_node, k_exec, _, _ = key
+                if k_stream != stream_id or k_node != node_id or k_exec != execution_id:
+                    continue
+            to_flush.append(key)
+
+        for key in to_flush:
+            evt = self._pending_output_snapshots.pop(key)
+            try:
+                line = json.dumps(evt.to_dict(), default=str)
+                self._session_log.write(line + "\n")
+            except Exception:
+                pass
+
+        if to_flush:
+            try:
+                self._session_log.flush()
+            except Exception:
+                pass
 
     def subscribe(
         self,
@@ -305,6 +446,19 @@ class EventBus:
         Args:
             event: Event to publish
         """
+        # Apply iteration offset at the source so ALL consumers (SSE subscribers,
+        # event history, session log) see the same monotonically increasing
+        # iteration values.  Without this, live SSE would use raw iterations
+        # while events.jsonl would use offset iterations, causing ID collisions
+        # on the frontend when replaying after cold resume.
+        if (
+            self._session_log_iteration_offset
+            and isinstance(event.data, dict)
+            and "iteration" in event.data
+        ):
+            offset = self._session_log_iteration_offset
+            event.data = {**event.data, "iteration": event.data["iteration"] + offset}
+
         # Add to history
         async with self._lock:
             self._event_history.append(event)
@@ -325,6 +479,15 @@ class EventBus:
                 except Exception:
                     pass  # never break event delivery
 
+        # Per-session persistent log (always-on when set_session_log was called).
+        # Streaming deltas are coalesced: client_output_delta and llm_text_delta
+        # are accumulated and flushed as a single snapshot event on llm_turn_complete.
+        if self._session_log is not None:
+            try:
+                self._write_session_log_event(event)
+            except Exception:
+                pass  # never break event delivery
+
         # Find matching subscriptions
         matching_handlers: list[EventHandler] = []
 
@@ -385,6 +548,7 @@ class EventBus:
         execution_id: str,
         input_data: dict[str, Any] | None = None,
         correlation_id: str | None = None,
+        run_id: str | None = None,
     ) -> None:
         """Emit execution started event."""
         await self.publish(
@@ -394,6 +558,7 @@ class EventBus:
                 execution_id=execution_id,
                 data={"input": input_data or {}},
                 correlation_id=correlation_id,
+                run_id=run_id,
             )
         )
 
@@ -403,6 +568,7 @@ class EventBus:
         execution_id: str,
         output: dict[str, Any] | None = None,
         correlation_id: str | None = None,
+        run_id: str | None = None,
     ) -> None:
         """Emit execution completed event."""
         await self.publish(
@@ -412,6 +578,7 @@ class EventBus:
                 execution_id=execution_id,
                 data={"output": output or {}},
                 correlation_id=correlation_id,
+                run_id=run_id,
             )
         )
 
@@ -421,6 +588,7 @@ class EventBus:
         execution_id: str,
         error: str,
         correlation_id: str | None = None,
+        run_id: str | None = None,
     ) -> None:
         """Emit execution failed event."""
         await self.publish(
@@ -430,6 +598,7 @@ class EventBus:
                 execution_id=execution_id,
                 data={"error": error},
                 correlation_id=correlation_id,
+                run_id=run_id,
             )
         )
 
@@ -521,15 +690,19 @@ class EventBus:
         node_id: str,
         iteration: int,
         execution_id: str | None = None,
+        extra_data: dict[str, Any] | None = None,
     ) -> None:
         """Emit node loop iteration event."""
+        data: dict[str, Any] = {"iteration": iteration}
+        if extra_data:
+            data.update(extra_data)
         await self.publish(
             AgentEvent(
                 type=EventType.NODE_LOOP_ITERATION,
                 stream_id=stream_id,
                 node_id=node_id,
                 execution_id=execution_id,
-                data={"iteration": iteration},
+                data=data,
             )
         )
 
@@ -578,6 +751,7 @@ class EventBus:
         content: str,
         snapshot: str,
         execution_id: str | None = None,
+        inner_turn: int = 0,
     ) -> None:
         """Emit LLM text delta event."""
         await self.publish(
@@ -586,7 +760,7 @@ class EventBus:
                 stream_id=stream_id,
                 node_id=node_id,
                 execution_id=execution_id,
-                data={"content": content, "snapshot": snapshot},
+                data={"content": content, "snapshot": snapshot, "inner_turn": inner_turn},
             )
         )
 
@@ -616,6 +790,7 @@ class EventBus:
         model: str,
         input_tokens: int,
         output_tokens: int,
+        cached_tokens: int = 0,
         execution_id: str | None = None,
         iteration: int | None = None,
     ) -> None:
@@ -625,6 +800,7 @@ class EventBus:
             "model": model,
             "input_tokens": input_tokens,
             "output_tokens": output_tokens,
+            "cached_tokens": cached_tokens,
         }
         if iteration is not None:
             data["iteration"] = iteration
@@ -700,9 +876,10 @@ class EventBus:
         snapshot: str,
         execution_id: str | None = None,
         iteration: int | None = None,
+        inner_turn: int = 0,
     ) -> None:
         """Emit client output delta event (client_facing=True nodes)."""
-        data: dict = {"content": content, "snapshot": snapshot}
+        data: dict = {"content": content, "snapshot": snapshot, "inner_turn": inner_turn}
         if iteration is not None:
             data["iteration"] = iteration
         await self.publish(
@@ -722,16 +899,23 @@ class EventBus:
         prompt: str = "",
         execution_id: str | None = None,
         options: list[str] | None = None,
+        questions: list[dict] | None = None,
     ) -> None:
         """Emit client input requested event (client_facing=True nodes).
 
         Args:
             options: Optional predefined choices for the user (1-3 items).
-                     The frontend appends an "Other" free-text option automatically.
+                     The frontend appends an "Other" free-text option
+                     automatically.
+            questions: Optional list of question dicts for multi-question
+                       batches (from ask_user_multiple). Each dict has id,
+                       prompt, and optional options.
         """
         data: dict[str, Any] = {"prompt": prompt}
         if options:
             data["options"] = options
+        if questions:
+            data["questions"] = questions
         await self.publish(
             AgentEvent(
                 type=EventType.CLIENT_INPUT_REQUESTED,
@@ -994,7 +1178,7 @@ class EventBus:
         ticket: dict,
         execution_id: str | None = None,
     ) -> None:
-        """Emitted by health judge when worker shows a degradation pattern."""
+        """Emitted when worker shows a degradation pattern."""
         await self.publish(
             AgentEvent(
                 type=EventType.WORKER_ESCALATION_TICKET,

core/framework/runtime/execution_stream.py

@@ -127,6 +127,7 @@ class ExecutionContext:
     input_data: dict[str, Any]
     isolation_level: IsolationLevel
     session_state: dict[str, Any] | None = None  # For resuming from pause
+    run_id: str | None = None  # Unique ID per trigger() invocation
     started_at: datetime = field(default_factory=datetime.now)
     completed_at: datetime | None = None
     status: str = "pending"  # pending, running, completed, failed, paused
@@ -185,6 +186,8 @@ class ExecutionStream:
         accounts_prompt: str = "",
         accounts_data: list[dict] | None = None,
         tool_provider_map: dict[str, str] | None = None,
+        skills_catalog_prompt: str = "",
+        protocols_prompt: str = "",
     ):
         """
         Initialize execution stream.
@@ -208,6 +211,8 @@ class ExecutionStream:
             accounts_prompt: Connected accounts block for system prompt injection
             accounts_data: Raw account data for per-node prompt generation
             tool_provider_map: Tool name to provider name mapping for account routing
+            skills_catalog_prompt: Available skills catalog for system prompt
+            protocols_prompt: Default skill operational protocols for system prompt
         """
         self.stream_id = stream_id
         self.entry_spec = entry_spec
@@ -229,6 +234,21 @@ class ExecutionStream:
         self._accounts_prompt = accounts_prompt
         self._accounts_data = accounts_data
         self._tool_provider_map = tool_provider_map
+        self._skills_catalog_prompt = skills_catalog_prompt
+        self._protocols_prompt = protocols_prompt
+
+        _es_logger = logging.getLogger(__name__)
+        if protocols_prompt:
+            _es_logger.info(
+                "ExecutionStream[%s] received protocols_prompt (%d chars)",
+                stream_id,
+                len(protocols_prompt),
+            )
+        else:
+            _es_logger.warning(
+                "ExecutionStream[%s] received EMPTY protocols_prompt",
+                stream_id,
+            )
 
         # Create stream-scoped runtime
         self._runtime = StreamRuntime(
@@ -425,11 +445,36 @@ class ExecutionStream:
                 return True
         return False
 
+    async def inject_trigger(
+        self,
+        node_id: str,
+        trigger: Any,
+    ) -> bool:
+        """Inject a trigger event into a running queen EventLoopNode.
+
+        Searches active executors for a node matching ``node_id`` and calls
+        its ``inject_trigger()`` method to wake the queen.
+
+        Args:
+            node_id: The queen EventLoopNode ID.
+            trigger: A ``TriggerEvent`` instance (typed as Any to avoid
+                circular imports with graph layer).
+
+        Returns True if the trigger was delivered, False otherwise.
+        """
+        for executor in self._active_executors.values():
+            node = executor.node_registry.get(node_id)
+            if node is not None and hasattr(node, "inject_trigger"):
+                await node.inject_trigger(trigger)
+                return True
+        return False
+
     async def execute(
         self,
         input_data: dict[str, Any],
         correlation_id: str | None = None,
         session_state: dict[str, Any] | None = None,
+        run_id: str | None = None,
     ) -> str:
         """
         Queue an execution and return its ID.
@@ -440,6 +485,7 @@ class ExecutionStream:
             input_data: Input data for this execution
             correlation_id: Optional ID to correlate related executions
             session_state: Optional session state to resume from (with paused_at, memory)
+            run_id: Unique ID for this trigger invocation (for run dividers)
 
         Returns:
             Execution ID for tracking
@@ -500,6 +546,7 @@ class ExecutionStream:
             input_data=input_data,
             isolation_level=self.entry_spec.get_isolation_level(),
             session_state=session_state,
+            run_id=run_id,
         )
 
         async with self._lock:
@@ -575,7 +622,9 @@ class ExecutionStream:
                         execution_id=execution_id,
                         input_data=ctx.input_data,
                         correlation_id=ctx.correlation_id,
+                        run_id=ctx.run_id,
                     )
+                self._write_run_event(execution_id, ctx.run_id, "run_started")
 
                 # Create execution-scoped memory
                 self._state_manager.create_memory(
@@ -645,6 +694,8 @@ class ExecutionStream:
                         accounts_prompt=self._accounts_prompt,
                         accounts_data=self._accounts_data,
                         tool_provider_map=self._tool_provider_map,
+                        skills_catalog_prompt=self._skills_catalog_prompt,
+                        protocols_prompt=self._protocols_prompt,
                     )
                     # Track executor so inject_input() can reach EventLoopNode instances
                     self._active_executors[execution_id] = executor
@@ -740,6 +791,7 @@ class ExecutionStream:
                             execution_id=execution_id,
                             output=result.output,
                             correlation_id=ctx.correlation_id,
+                            run_id=ctx.run_id,
                         )
                     elif result.paused_at:
                         # The executor returns paused_at on CancelledError but
@@ -757,8 +809,22 @@ class ExecutionStream:
                             execution_id=execution_id,
                             error=result.error or "Unknown error",
                             correlation_id=ctx.correlation_id,
+                            run_id=ctx.run_id,
                         )
 
+                # Write run event for historical restoration
+                if result.success:
+                    self._write_run_event(execution_id, ctx.run_id, "run_completed")
+                elif result.paused_at:
+                    self._write_run_event(execution_id, ctx.run_id, "run_paused")
+                else:
+                    self._write_run_event(
+                        execution_id,
+                        ctx.run_id,
+                        "run_failed",
+                        {"error": result.error or "Unknown error"},
+                    )
+
                 logger.debug(f"Execution {execution_id} completed: success={result.success}")
 
             except asyncio.CancelledError:
@@ -818,8 +884,10 @@ class ExecutionStream:
                             execution_id=execution_id,
                             error=cancel_reason,
                             correlation_id=ctx.correlation_id,
+                            run_id=ctx.run_id,
                         )
 
+                self._write_run_event(execution_id, ctx.run_id, "run_cancelled")
                 # Don't re-raise - we've handled it and saved state
 
             except Exception as e:
@@ -856,7 +924,9 @@ class ExecutionStream:
                         execution_id=execution_id,
                         error=str(e),
                         correlation_id=ctx.correlation_id,
+                        run_id=ctx.run_id,
                     )
+                self._write_run_event(execution_id, ctx.run_id, "run_failed", {"error": str(e)})
 
             finally:
                 # Clean up state
@@ -872,6 +942,36 @@ class ExecutionStream:
                     self._completion_events.pop(execution_id, None)
                     self._execution_tasks.pop(execution_id, None)
 
+    def _write_run_event(
+        self,
+        execution_id: str,
+        run_id: str | None,
+        event: str,
+        extra: dict[str, Any] | None = None,
+    ) -> None:
+        """Append a run lifecycle event to runs.jsonl for historical restoration."""
+        if not self._session_store or not run_id:
+            return
+        import json as _json
+
+        session_dir = self._session_store.get_session_path(execution_id)
+        runs_file = session_dir / "runs.jsonl"
+        now = datetime.now()
+        record = {
+            "run_id": run_id,
+            "event": event,
+            "timestamp": now.isoformat(),
+            "created_at": now.timestamp(),
+        }
+        if extra:
+            record.update(extra)
+        try:
+            runs_file.parent.mkdir(parents=True, exist_ok=True)
+            with open(runs_file, "a", encoding="utf-8") as f:
+                f.write(_json.dumps(record) + "\n")
+        except OSError:
+            pass  # Non-critical — don't break execution
+
     async def _write_session_state(
         self,
         execution_id: str,
@@ -978,8 +1078,8 @@ class ExecutionStream:
     def _create_modified_graph(self) -> "GraphSpec":
         """Create a graph with the entry point overridden.
 
-        Preserves the original graph's entry_points and async_entry_points
-        so that validation correctly considers ALL entry nodes reachable.
+        Preserves the original graph's entry_points so that validation
+        correctly considers ALL entry nodes reachable.
         Each stream only executes from its own entry_node, but the full
         graph must validate with all entry points accounted for.
         """
@@ -1004,7 +1104,6 @@ class ExecutionStream:
             version=self.graph.version,
             entry_node=self.entry_spec.entry_node,  # Use our entry point
             entry_points=merged_entry_points,
-            async_entry_points=self.graph.async_entry_points,
             terminal_nodes=self.graph.terminal_nodes,
             pause_nodes=self.graph.pause_nodes,
             nodes=self.graph.nodes,

core/framework/runtime/runtime_log_store.py

@@ -47,25 +47,34 @@ class RuntimeLogStore:
         self._base_path = base_path
         # Note: _runs_dir is determined per-run_id by _get_run_dir()
 
+    def _session_logs_dir(self, run_id: str) -> Path:
+        """Return the unified session-backed logs directory for a run ID."""
+        is_runtime_logs = self._base_path.name == "runtime_logs"
+        root = self._base_path.parent if is_runtime_logs else self._base_path
+        return root / "sessions" / run_id / "logs"
+
+    def _legacy_run_dir(self, run_id: str) -> Path:
+        """Return the deprecated standalone runs directory for a run ID."""
+        return self._base_path / "runs" / run_id
+
     def _get_run_dir(self, run_id: str) -> Path:
         """Determine run directory path based on run_id format.
 
-        - New format (session_*): {storage_root}/sessions/{run_id}/logs/
+        - Session-backed runs: {storage_root}/sessions/{run_id}/logs/
         - Old format (anything else): {base_path}/runs/{run_id}/ (deprecated)
         """
-        if run_id.startswith("session_"):
-            is_runtime_logs = self._base_path.name == "runtime_logs"
-            root = self._base_path.parent if is_runtime_logs else self._base_path
-            return root / "sessions" / run_id / "logs"
+        session_run_dir = self._session_logs_dir(run_id)
+        if session_run_dir.exists() or run_id.startswith("session_"):
+            return session_run_dir
         import warnings
 
         warnings.warn(
             f"Reading logs from deprecated location for run_id={run_id}. "
-            "New sessions use unified storage at sessions/session_*/logs/",
+            "New sessions use unified storage at sessions/<session_id>/logs/",
             DeprecationWarning,
             stacklevel=3,
         )
-        return self._base_path / "runs" / run_id
+        return self._legacy_run_dir(run_id)
 
     # -------------------------------------------------------------------
     # Incremental write (sync — called from locked sections)
@@ -76,6 +85,10 @@ class RuntimeLogStore:
         run_dir = self._get_run_dir(run_id)
         run_dir.mkdir(parents=True, exist_ok=True)
 
+    def ensure_session_run_dir(self, run_id: str) -> None:
+        """Create the unified session-backed log directory immediately."""
+        self._session_logs_dir(run_id).mkdir(parents=True, exist_ok=True)
+
     def append_step(self, run_id: str, step: NodeStepLog) -> None:
         """Append one JSONL line to tool_logs.jsonl. Sync."""
         path = self._get_run_dir(run_id) / "tool_logs.jsonl"
@@ -200,17 +213,17 @@ class RuntimeLogStore:
         run_ids = []
 
         # Scan new location: base_path/sessions/{session_id}/logs/
-        # Determine the correct base path for sessions
         is_runtime_logs = self._base_path.name == "runtime_logs"
         root = self._base_path.parent if is_runtime_logs else self._base_path
         sessions_dir = root / "sessions"
 
         if sessions_dir.exists():
             for session_dir in sessions_dir.iterdir():
-                if session_dir.is_dir() and session_dir.name.startswith("session_"):
-                    logs_dir = session_dir / "logs"
-                    if logs_dir.exists() and logs_dir.is_dir():
-                        run_ids.append(session_dir.name)
+                if not session_dir.is_dir():
+                    continue
+                logs_dir = session_dir / "logs"
+                if logs_dir.exists() and logs_dir.is_dir():
+                    run_ids.append(session_dir.name)
 
         # Scan old location: base_path/runs/ (deprecated)
         old_runs_dir = self._base_path / "runs"

core/framework/runtime/runtime_logger.py

@@ -66,15 +66,16 @@ class RuntimeLogger:
         """
         if session_id:
             self._run_id = session_id
+            self._store.ensure_session_run_dir(self._run_id)
         else:
             ts = datetime.now(UTC).strftime("%Y%m%dT%H%M%S")
             short_uuid = uuid.uuid4().hex[:8]
             self._run_id = f"{ts}_{short_uuid}"
+            self._store.ensure_run_dir(self._run_id)
 
         self._goal_id = goal_id
         self._started_at = datetime.now(UTC).isoformat()
         self._logged_node_ids = set()
-        self._store.ensure_run_dir(self._run_id)
         return self._run_id
 
     def log_step(

core/framework/runtime/tests/test_agent_runtime.py

@@ -17,7 +17,7 @@ from pathlib import Path
 import pytest
 
 from framework.graph import Goal
-from framework.graph.edge import AsyncEntryPointSpec, EdgeCondition, EdgeSpec, GraphSpec
+from framework.graph.edge import EdgeCondition, EdgeSpec, GraphSpec
 from framework.graph.goal import Constraint, SuccessCriterion
 from framework.graph.node import NodeSpec
 from framework.runtime.agent_runtime import AgentRuntime, create_agent_runtime
@@ -101,30 +101,12 @@ def sample_graph():
         ),
     ]
 
-    async_entry_points = [
-        AsyncEntryPointSpec(
-            id="webhook",
-            name="Webhook Handler",
-            entry_node="process-webhook",
-            trigger_type="webhook",
-            isolation_level="shared",
-        ),
-        AsyncEntryPointSpec(
-            id="api",
-            name="API Handler",
-            entry_node="process-api",
-            trigger_type="api",
-            isolation_level="shared",
-        ),
-    ]
-
     return GraphSpec(
         id="test-graph",
         goal_id="test-goal",
         version="1.0.0",
         entry_node="process-webhook",
         entry_points={"start": "process-webhook"},
-        async_entry_points=async_entry_points,
         terminal_nodes=["complete"],
         pause_nodes=[],
         nodes=nodes,
@@ -504,108 +486,6 @@ class TestAgentRuntime:
 # === GraphSpec Validation Tests ===
 
 
-class TestGraphSpecValidation:
-    """Tests for GraphSpec with async_entry_points."""
-
-    def test_has_async_entry_points(self, sample_graph):
-        """Test checking for async entry points."""
-        assert sample_graph.has_async_entry_points() is True
-
-        # Graph without async entry points
-        simple_graph = GraphSpec(
-            id="simple",
-            goal_id="goal",
-            entry_node="start",
-            nodes=[],
-            edges=[],
-        )
-        assert simple_graph.has_async_entry_points() is False
-
-    def test_get_async_entry_point(self, sample_graph):
-        """Test getting async entry point by ID."""
-        ep = sample_graph.get_async_entry_point("webhook")
-        assert ep is not None
-        assert ep.id == "webhook"
-        assert ep.entry_node == "process-webhook"
-
-        ep_not_found = sample_graph.get_async_entry_point("nonexistent")
-        assert ep_not_found is None
-
-    def test_validate_async_entry_points(self):
-        """Test validation catches async entry point errors."""
-        nodes = [
-            NodeSpec(
-                id="valid-node",
-                name="Valid Node",
-                description="A valid node",
-                node_type="event_loop",
-                input_keys=[],
-                output_keys=[],
-            ),
-        ]
-
-        # Invalid entry node
-        graph = GraphSpec(
-            id="test",
-            goal_id="goal",
-            entry_node="valid-node",
-            async_entry_points=[
-                AsyncEntryPointSpec(
-                    id="invalid",
-                    name="Invalid",
-                    entry_node="nonexistent-node",
-                    trigger_type="webhook",
-                ),
-            ],
-            nodes=nodes,
-            edges=[],
-        )
-
-        errors = graph.validate()["errors"]
-        assert any("nonexistent-node" in e for e in errors)
-
-        # Invalid isolation level
-        graph2 = GraphSpec(
-            id="test",
-            goal_id="goal",
-            entry_node="valid-node",
-            async_entry_points=[
-                AsyncEntryPointSpec(
-                    id="bad-isolation",
-                    name="Bad Isolation",
-                    entry_node="valid-node",
-                    trigger_type="webhook",
-                    isolation_level="invalid",
-                ),
-            ],
-            nodes=nodes,
-            edges=[],
-        )
-
-        errors2 = graph2.validate()["errors"]
-        assert any("isolation_level" in e for e in errors2)
-
-        # Invalid trigger type
-        graph3 = GraphSpec(
-            id="test",
-            goal_id="goal",
-            entry_node="valid-node",
-            async_entry_points=[
-                AsyncEntryPointSpec(
-                    id="bad-trigger",
-                    name="Bad Trigger",
-                    entry_node="valid-node",
-                    trigger_type="invalid_trigger",
-                ),
-            ],
-            nodes=nodes,
-            edges=[],
-        )
-
-        errors3 = graph3.validate()["errors"]
-        assert any("trigger_type" in e for e in errors3)
-
-
 # === Integration Tests ===
 
 

core/framework/runtime/tests/test_runtime_logging_paths.py

@@ -0,0 +1,29 @@
+"""Tests for custom session-backed runtime logging paths."""
+
+from pathlib import Path
+from unittest.mock import MagicMock
+
+from framework.graph.executor import GraphExecutor
+from framework.runtime.runtime_log_store import RuntimeLogStore
+from framework.runtime.runtime_logger import RuntimeLogger
+
+
+def test_graph_executor_uses_custom_session_dir_name_for_runtime_logs():
+    executor = GraphExecutor(
+        runtime=MagicMock(),
+        storage_path=Path("/tmp/test-agent/sessions/my-custom-session"),
+    )
+
+    assert executor._get_runtime_log_session_id() == "my-custom-session"
+
+
+def test_runtime_logger_creates_session_log_dir_for_custom_session_id(tmp_path):
+    base = tmp_path / ".hive" / "agents" / "test_agent"
+    base.mkdir(parents=True)
+    store = RuntimeLogStore(base)
+    logger = RuntimeLogger(store=store, agent_id="test-agent")
+
+    run_id = logger.start_run(goal_id="goal-1", session_id="my-custom-session")
+
+    assert run_id == "my-custom-session"
+    assert (base / "sessions" / "my-custom-session" / "logs").is_dir()

core/framework/runtime/tests/test_webhook_server.py

@@ -483,7 +483,6 @@ class TestEventDrivenEntryPoints:
             version="1.0.0",
             entry_node="process-event",
             entry_points={"start": "process-event"},
-            async_entry_points=[],
             terminal_nodes=[],
             pause_nodes=[],
             nodes=nodes,

core/framework/runtime/triggers.py

@@ -0,0 +1,22 @@
+"""Trigger definitions for queen-level heartbeats (timers, webhooks)."""
+
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+from typing import Any
+
+
+@dataclass
+class TriggerDefinition:
+    """A registered trigger that can be activated on the queen runtime.
+
+    Trigger *definitions* come from the worker's ``triggers.json``.
+    Activation state is per-session (persisted in ``SessionState.active_triggers``).
+    """
+
+    id: str
+    trigger_type: str  # "timer" | "webhook"
+    trigger_config: dict[str, Any] = field(default_factory=dict)
+    description: str = ""
+    task: str = ""
+    active: bool = False

core/framework/schemas/session_state.py

@@ -144,6 +144,13 @@ class SessionState(BaseModel):
     checkpoint_enabled: bool = False
     latest_checkpoint_id: str | None = None
 
+    # Trigger activation state (IDs of triggers the queen/user turned on)
+    active_triggers: list[str] = Field(default_factory=list)
+    # Per-trigger task strings (user overrides, keyed by trigger ID)
+    trigger_tasks: dict[str, str] = Field(default_factory=dict)
+    # True after first successful worker execution (gates trigger delivery on restart)
+    worker_configured: bool = Field(default=False)
+
     model_config = {"extra": "allow"}
 
     @computed_field

core/framework/server/app.py

@@ -94,6 +94,29 @@ def sessions_dir(session: Session) -> Path:
     return Path.home() / ".hive" / "agents" / agent_name / "sessions"
 
 
+def cold_sessions_dir(session_id: str) -> Path | None:
+    """Resolve the worker sessions directory from disk for a cold/stopped session.
+
+    Reads agent_path from the queen session's meta.json to find the agent name,
+    then returns ~/.hive/agents/{agent_name}/sessions/.
+    Returns None if meta.json is missing or has no agent_path.
+    """
+    import json
+
+    meta_path = Path.home() / ".hive" / "queen" / "session" / session_id / "meta.json"
+    if not meta_path.exists():
+        return None
+    try:
+        meta = json.loads(meta_path.read_text(encoding="utf-8"))
+        agent_path = meta.get("agent_path")
+        if not agent_path:
+            return None
+        agent_name = Path(agent_path).name
+        return Path.home() / ".hive" / "agents" / agent_name / "sessions"
+    except (json.JSONDecodeError, OSError):
+        return None
+
+
 # Allowed CORS origins (localhost on any port)
 _CORS_ORIGINS = {"http://localhost", "http://127.0.0.1"}
 

core/framework/server/queen_orchestrator.py

@@ -69,6 +69,7 @@ async def create_queen(
         QueenPhaseState,
         register_queen_lifecycle_tools,
     )
+    from framework.tools.queen_memory_tools import register_queen_memory_tools
 
     hive_home = Path.home() / ".hive"
 
@@ -90,6 +91,28 @@ async def create_queen(
     phase_state = QueenPhaseState(phase=initial_phase, event_bus=session.event_bus)
     session.phase_state = phase_state
 
+    # ---- Track ask rounds during planning ----------------------------
+    # Increment planning_ask_rounds each time the queen requests user
+    # input (ask_user or ask_user_multiple) while in the planning phase.
+    async def _track_planning_asks(event: AgentEvent) -> None:
+        if phase_state.phase != "planning":
+            return
+        # Only count explicit ask_user / ask_user_multiple calls, not
+        # auto-block (text-only turns emit CLIENT_INPUT_REQUESTED with
+        # an empty prompt and no options/questions).
+        data = event.data or {}
+        has_prompt = bool(data.get("prompt"))
+        has_questions = bool(data.get("questions"))
+        has_options = bool(data.get("options"))
+        if has_prompt or has_questions or has_options:
+            phase_state.planning_ask_rounds += 1
+
+    session.event_bus.subscribe(
+        [EventType.CLIENT_INPUT_REQUESTED],
+        _track_planning_asks,
+        filter_stream="queen",
+    )
+
     # ---- Lifecycle tools (always registered) --------------------------
     register_queen_lifecycle_tools(
         queen_registry,
@@ -100,6 +123,9 @@ async def create_queen(
         phase_state=phase_state,
     )
 
+    # ---- Episodic memory tools (always registered) ---------------------
+    register_queen_memory_tools(queen_registry)
+
     # ---- Monitoring tools (only when worker is loaded) ----------------
     if session.worker_runtime:
         from framework.tools.worker_monitoring_tools import register_worker_monitoring_tools
@@ -110,6 +136,7 @@ async def create_queen(
             session.worker_path,
             stream_id="queen",
             worker_graph_id=session.worker_runtime._graph_id,
+            default_session_id=session.id,
         )
 
     queen_tools = list(queen_registry.get_tools().values())
@@ -149,7 +176,8 @@ async def create_queen(
         worker_identity = (
             "\n\n# Worker Profile\n"
             "No worker agent loaded. You are operating independently.\n"
-            "Handle all tasks directly using your coding tools."
+            "Design or build the agent to solve the user's problem "
+            "according to your current phase."
         )
 
     _planning_body = (
@@ -192,6 +220,16 @@ async def create_queen(
         + worker_identity
     )
 
+    # ---- Default skill protocols -------------------------------------
+    try:
+        from framework.skills.manager import SkillsManager
+
+        _queen_skills_mgr = SkillsManager()
+        _queen_skills_mgr.load()
+        phase_state.protocols_prompt = _queen_skills_mgr.protocols_prompt
+    except Exception:
+        logger.debug("Queen skill loading failed (non-fatal)", exc_info=True)
+
     # ---- Persona hook ------------------------------------------------
     _session_llm = session.llm
     _session_event_bus = session.event_bus
@@ -252,6 +290,7 @@ async def create_queen(
                 execution_id=session.id,
                 dynamic_tools_provider=phase_state.get_current_tools,
                 dynamic_prompt_provider=phase_state.get_current_prompt,
+                iteration_metadata_provider=lambda: {"phase": phase_state.phase},
             )
             session.queen_executor = executor
 
@@ -269,6 +308,8 @@ async def create_queen(
                     return
                 if phase_state.phase == "running":
                     if event.type == EventType.EXECUTION_COMPLETED:
+                        # Mark worker as configured after first successful run
+                        session.worker_configured = True
                         output = event.data.get("output", {})
                         output_summary = ""
                         if output:

core/framework/server/routes_credentials.py

@@ -103,7 +103,9 @@ async def handle_delete_credential(request: web.Request) -> web.Response:
     if credential_id == "aden_api_key":
         from framework.credentials.key_storage import delete_aden_api_key
 
-        delete_aden_api_key()
+        deleted = delete_aden_api_key()
+        if not deleted:
+            return web.json_response({"error": "Credential 'aden_api_key' not found"}, status=404)
         return web.json_response({"deleted": True})
 
     store = _get_store(request)
@@ -178,7 +180,10 @@ async def handle_check_agent(request: web.Request) -> web.Response:
         )
     except Exception as e:
         logger.exception(f"Error checking agent credentials: {e}")
-        return web.json_response({"error": str(e)}, status=500)
+        return web.json_response(
+            {"error": "Internal server error while checking credentials"},
+            status=500,
+        )
 
 
 def _status_to_dict(c) -> dict:

core/framework/server/routes_events.py

@@ -6,7 +6,7 @@ import logging
 from aiohttp import web
 from aiohttp.client_exceptions import ClientConnectionResetError as _AiohttpConnReset
 
-from framework.runtime.event_bus import EventType
+from framework.runtime.event_bus import AgentEvent, EventType
 from framework.server.app import resolve_session
 
 logger = logging.getLogger(__name__)
@@ -15,6 +15,7 @@ logger = logging.getLogger(__name__)
 DEFAULT_EVENT_TYPES = [
     EventType.CLIENT_OUTPUT_DELTA,
     EventType.CLIENT_INPUT_REQUESTED,
+    EventType.CLIENT_INPUT_RECEIVED,
     EventType.LLM_TEXT_DELTA,
     EventType.TOOL_CALL_STARTED,
     EventType.TOOL_CALL_COMPLETED,
@@ -40,6 +41,13 @@ DEFAULT_EVENT_TYPES = [
     EventType.CREDENTIALS_REQUIRED,
     EventType.SUBAGENT_REPORT,
     EventType.QUEEN_PHASE_CHANGED,
+    EventType.TRIGGER_AVAILABLE,
+    EventType.TRIGGER_ACTIVATED,
+    EventType.TRIGGER_DEACTIVATED,
+    EventType.TRIGGER_FIRED,
+    EventType.TRIGGER_REMOVED,
+    EventType.TRIGGER_UPDATED,
+    EventType.DRAFT_GRAPH_UPDATED,
 ]
 
 # Keepalive interval in seconds
@@ -89,6 +97,7 @@ async def handle_events(request: web.Request) -> web.StreamResponse:
         "execution_failed",
         "execution_paused",
         "client_input_requested",
+        "client_input_received",
         "node_loop_iteration",
         "node_loop_started",
         "credentials_required",
@@ -142,6 +151,7 @@ async def handle_events(request: web.Request) -> web.StreamResponse:
         EventType.CLIENT_OUTPUT_DELTA.value,
         EventType.EXECUTION_STARTED.value,
         EventType.CLIENT_INPUT_REQUESTED.value,
+        EventType.CLIENT_INPUT_RECEIVED.value,
     }
     event_type_values = {et.value for et in event_types}
     replay_types = _REPLAY_TYPES & event_type_values
@@ -156,6 +166,54 @@ async def handle_events(request: web.Request) -> web.StreamResponse:
     if replayed:
         logger.info("SSE replayed %d buffered events for session='%s'", replayed, session.id)
 
+    # Inject a live-status snapshot so the frontend knows which nodes are
+    # currently running.  This covers the case where the user navigated away
+    # and back — the localStorage snapshot is stale, and the ring-buffer
+    # replay may not include the original node_loop_started events.
+    worker_runtime = getattr(session, "worker_runtime", None)
+    if worker_runtime and getattr(worker_runtime, "is_running", False):
+        try:
+            for stream_info in worker_runtime.get_active_streams():
+                graph_id = stream_info.get("graph_id")
+                stream_id = stream_info.get("stream_id", "default")
+                for exec_id in stream_info.get("active_execution_ids", []):
+                    # Synthesize execution_started so frontend sets workerRunState
+                    synth_exec = AgentEvent(
+                        type=EventType.EXECUTION_STARTED,
+                        stream_id=stream_id,
+                        execution_id=exec_id,
+                        graph_id=graph_id,
+                        data={"synthetic": True},
+                    ).to_dict()
+                    try:
+                        queue.put_nowait(synth_exec)
+                    except asyncio.QueueFull:
+                        pass
+
+                # Find the currently executing node via the executor
+                for _gid, reg in worker_runtime._graphs.items():
+                    if _gid != graph_id:
+                        continue
+                    for _ep_id, stream in reg.streams.items():
+                        for exec_id, executor in stream._active_executors.items():
+                            current = getattr(executor, "current_node_id", None)
+                            if current:
+                                synth_node = AgentEvent(
+                                    type=EventType.NODE_LOOP_STARTED,
+                                    stream_id=stream_id,
+                                    node_id=current,
+                                    execution_id=exec_id,
+                                    graph_id=graph_id,
+                                    data={"synthetic": True},
+                                ).to_dict()
+                                try:
+                                    queue.put_nowait(synth_node)
+                                except asyncio.QueueFull:
+                                    pass
+            logger.info("SSE injected live-status snapshot for session='%s'", session.id)
+        except Exception:
+            logger.debug("Failed to inject live-status snapshot", exc_info=True)
+
     event_count = 0
     close_reason = "unknown"
     try:

core/framework/server/routes_execution.py

@@ -125,6 +125,18 @@ async def handle_chat(request: web.Request) -> web.Response:
         node = queen_executor.node_registry.get("queen")
         if node is not None and hasattr(node, "inject_event"):
             await node.inject_event(message, is_client_input=True)
+            # Publish to EventBus so the session event log captures user messages
+            from framework.runtime.event_bus import AgentEvent, EventType
+
+            await session.event_bus.publish(
+                AgentEvent(
+                    type=EventType.CLIENT_INPUT_RECEIVED,
+                    stream_id="queen",
+                    node_id="queen",
+                    execution_id=session.id,
+                    data={"content": message},
+                )
+            )
             return web.json_response(
                 {
                     "status": "queen",

core/framework/server/routes_graphs.py

@@ -2,6 +2,7 @@
 
 import json
 import logging
+import time
 
 from aiohttp import web
 
@@ -116,6 +117,20 @@ async def handle_list_nodes(request: web.Request) -> web.Response:
         }
         for ep in reg.entry_points.values()
     ]
+    # Append triggers from triggers.json (stored on session)
+    for t in getattr(session, "available_triggers", {}).values():
+        entry = {
+            "id": t.id,
+            "name": t.description or t.id,
+            "entry_node": graph.entry_node,
+            "trigger_type": t.trigger_type,
+            "trigger_config": t.trigger_config,
+            "task": t.task,
+        }
+        mono = getattr(session, "trigger_next_fire", {}).get(t.id)
+        if mono is not None:
+            entry["next_fire_in"] = max(0.0, mono - time.monotonic())
+        entry_points.append(entry)
     return web.json_response(
         {
             "nodes": nodes,
@@ -234,8 +249,73 @@ async def handle_node_tools(request: web.Request) -> web.Response:
     return web.json_response({"tools": tools_out})
 
 
+async def handle_draft_graph(request: web.Request) -> web.Response:
+    """Return the current draft graph from planning phase (if any)."""
+    session, err = resolve_session(request)
+    if err:
+        return err
+
+    phase_state = getattr(session, "phase_state", None)
+    if phase_state is None or phase_state.draft_graph is None:
+        return web.json_response({"draft": None})
+
+    return web.json_response({"draft": phase_state.draft_graph})
+
+
+async def handle_flowchart_map(request: web.Request) -> web.Response:
+    """Return the flowchart→runtime node mapping and the original (pre-dissolution) draft.
+
+    Available after confirm_and_build() dissolves decision nodes, or loaded
+    from the agent's flowchart.json file, or synthesized from the runtime graph.
+    """
+    session, err = resolve_session(request)
+    if err:
+        return err
+
+    phase_state = getattr(session, "phase_state", None)
+
+    # Fast path: already in memory
+    if phase_state is not None and phase_state.original_draft_graph is not None:
+        return web.json_response(
+            {
+                "map": phase_state.flowchart_map,
+                "original_draft": phase_state.original_draft_graph,
+            }
+        )
+
+    # Try loading from flowchart.json in the agent folder
+    worker_path = getattr(session, "worker_path", None)
+    if worker_path is not None:
+        from pathlib import Path
+
+        target = Path(worker_path) / "flowchart.json"
+        if target.is_file():
+            try:
+                data = json.loads(target.read_text(encoding="utf-8"))
+                original_draft = data.get("original_draft")
+                fmap = data.get("flowchart_map")
+                # Cache in phase_state for future requests
+                if phase_state is not None and original_draft:
+                    phase_state.original_draft_graph = original_draft
+                    phase_state.flowchart_map = fmap
+                return web.json_response(
+                    {
+                        "map": fmap,
+                        "original_draft": original_draft,
+                    }
+                )
+            except Exception:
+                logger.warning("Failed to read flowchart.json from %s", worker_path)
+
+    return web.json_response({"map": None, "original_draft": None})
+
+
 def register_routes(app: web.Application) -> None:
     """Register graph/node inspection routes."""
+    # Draft graph (planning phase — visual only, no loaded worker required)
+    app.router.add_get("/api/sessions/{session_id}/draft-graph", handle_draft_graph)
+    # Flowchart map (post-dissolution — maps runtime nodes to original draft nodes)
+    app.router.add_get("/api/sessions/{session_id}/flowchart-map", handle_flowchart_map)
     # Session-primary routes
     app.router.add_get("/api/sessions/{session_id}/graphs/{graph_id}/nodes", handle_list_nodes)
     app.router.add_get(

core/framework/server/routes_sessions.py

@@ -9,8 +9,9 @@ Session-primary routes:
 - DELETE /api/sessions/{session_id}/worker           — unload worker from session
 - GET    /api/sessions/{session_id}/stats            — runtime statistics
 - GET    /api/sessions/{session_id}/entry-points     — list entry points
+- PATCH  /api/sessions/{session_id}/triggers/{id}   — update trigger task
 - GET    /api/sessions/{session_id}/graphs           — list graph IDs
-- GET    /api/sessions/{session_id}/queen-messages   — queen conversation history
+- GET    /api/sessions/{session_id}/events/history  — persisted eventbus log (for replay)
 
 Worker session browsing (persisted execution runs on disk):
 - GET    /api/sessions/{session_id}/worker-sessions                             — list
@@ -22,6 +23,8 @@ Worker session browsing (persisted execution runs on disk):
 
 """
 
+import asyncio
+import contextlib
 import json
 import logging
 import shutil
@@ -31,6 +34,7 @@ from pathlib import Path
 from aiohttp import web
 
 from framework.server.app import (
+    cold_sessions_dir,
     resolve_session,
     safe_path_segment,
     sessions_dir,
@@ -61,7 +65,9 @@ def _session_to_live_dict(session) -> dict:
         "loaded_at": session.loaded_at,
         "uptime_seconds": round(time.time() - session.loaded_at, 1),
         "intro_message": getattr(session.runner, "intro_message", "") or "",
-        "queen_phase": phase_state.phase if phase_state else "planning",
+        "queen_phase": phase_state.phase
+        if phase_state
+        else ("staging" if session.worker_runtime else "planning"),
     }
 
 
@@ -140,6 +146,7 @@ async def handle_create_session(request: web.Request) -> web.Response:
             session = await manager.create_session_with_worker(
                 agent_path,
                 agent_id=agent_id,
+                session_id=session_id,
                 model=model,
                 initial_prompt=initial_prompt,
                 queen_resume_from=queen_resume_from,
@@ -228,6 +235,22 @@ async def handle_get_live_session(request: web.Request) -> web.Response:
             }
             for ep in rt.get_entry_points()
         ]
+        # Append triggers from triggers.json (stored on session)
+        runner = getattr(session, "runner", None)
+        graph_entry = runner.graph.entry_node if runner else ""
+        for t in getattr(session, "available_triggers", {}).values():
+            entry = {
+                "id": t.id,
+                "name": t.description or t.id,
+                "entry_node": graph_entry,
+                "trigger_type": t.trigger_type,
+                "trigger_config": t.trigger_config,
+                "task": t.task,
+            }
+            mono = getattr(session, "trigger_next_fire", {}).get(t.id)
+            if mono is not None:
+                entry["next_fire_in"] = max(0.0, mono - time.monotonic())
+            data["entry_points"].append(entry)
         data["graphs"] = session.worker_runtime.list_graphs()
 
     return web.json_response(data)
@@ -351,23 +374,190 @@ async def handle_session_entry_points(request: web.Request) -> web.Response:
 
     rt = session.worker_runtime
     eps = rt.get_entry_points() if rt else []
+    entry_points = [
+        {
+            "id": ep.id,
+            "name": ep.name,
+            "entry_node": ep.entry_node,
+            "trigger_type": ep.trigger_type,
+            "trigger_config": ep.trigger_config,
+            **(
+                {"next_fire_in": nf}
+                if rt and (nf := rt.get_timer_next_fire_in(ep.id)) is not None
+                else {}
+            ),
+        }
+        for ep in eps
+    ]
+    # Append triggers from triggers.json (stored on session)
+    runner = getattr(session, "runner", None)
+    graph_entry = runner.graph.entry_node if runner else ""
+    for t in getattr(session, "available_triggers", {}).values():
+        entry = {
+            "id": t.id,
+            "name": t.description or t.id,
+            "entry_node": graph_entry,
+            "trigger_type": t.trigger_type,
+            "trigger_config": t.trigger_config,
+            "task": t.task,
+        }
+        mono = getattr(session, "trigger_next_fire", {}).get(t.id)
+        if mono is not None:
+            entry["next_fire_in"] = max(0.0, mono - time.monotonic())
+        entry_points.append(entry)
+    return web.json_response({"entry_points": entry_points})
+
+
+async def handle_update_trigger_task(request: web.Request) -> web.Response:
+    """PATCH /api/sessions/{session_id}/triggers/{trigger_id} — update trigger fields."""
+    session, err = resolve_session(request)
+    if err:
+        return err
+
+    trigger_id = request.match_info["trigger_id"]
+    available = getattr(session, "available_triggers", {})
+    tdef = available.get(trigger_id)
+    if tdef is None:
+        return web.json_response(
+            {"error": f"Trigger '{trigger_id}' not found"},
+            status=404,
+        )
+
+    try:
+        body = await request.json()
+    except Exception:
+        return web.json_response({"error": "Invalid JSON body"}, status=400)
+
+    updates: dict[str, object] = {}
+
+    if "task" in body:
+        task = body.get("task")
+        if not isinstance(task, str):
+            return web.json_response({"error": "'task' must be a string"}, status=400)
+        tdef.task = task
+        updates["task"] = tdef.task
+
+    trigger_config_update = body.get("trigger_config")
+    if trigger_config_update is not None:
+        if not isinstance(trigger_config_update, dict):
+            return web.json_response(
+                {"error": "'trigger_config' must be an object"},
+                status=400,
+            )
+        merged_trigger_config = dict(tdef.trigger_config)
+        merged_trigger_config.update(trigger_config_update)
+
+        if tdef.trigger_type == "timer":
+            cron_expr = merged_trigger_config.get("cron")
+            interval = merged_trigger_config.get("interval_minutes")
+            if cron_expr is not None and not isinstance(cron_expr, str):
+                return web.json_response(
+                    {"error": "'trigger_config.cron' must be a string"},
+                    status=400,
+                )
+            if cron_expr:
+                try:
+                    from croniter import croniter
+
+                    if not croniter.is_valid(cron_expr):
+                        return web.json_response(
+                            {"error": f"Invalid cron expression: {cron_expr}"},
+                            status=400,
+                        )
+                except ImportError:
+                    return web.json_response(
+                        {
+                            "error": (
+                                "croniter package not installed — cannot validate cron expression."
+                            )
+                        },
+                        status=500,
+                    )
+                merged_trigger_config.pop("interval_minutes", None)
+            elif interval is None:
+                return web.json_response(
+                    {
+                        "error": (
+                            "Timer trigger needs 'cron' or 'interval_minutes' in trigger_config."
+                        )
+                    },
+                    status=400,
+                )
+            elif not isinstance(interval, (int, float)) or interval <= 0:
+                return web.json_response(
+                    {"error": "'trigger_config.interval_minutes' must be > 0"},
+                    status=400,
+                )
+        tdef.trigger_config = merged_trigger_config
+        updates["trigger_config"] = tdef.trigger_config
+
+    if not updates:
+        return web.json_response(
+            {"error": "Provide at least one of 'task' or 'trigger_config'"},
+            status=400,
+        )
+
+    # Persist to session state and agent definition
+    from framework.tools.queen_lifecycle_tools import (
+        _persist_active_triggers,
+        _save_trigger_to_agent,
+        _start_trigger_timer,
+        _start_trigger_webhook,
+    )
+
+    if "trigger_config" in updates and trigger_id in getattr(session, "active_trigger_ids", set()):
+        task = session.active_timer_tasks.pop(trigger_id, None)
+        if task and not task.done():
+            task.cancel()
+            with contextlib.suppress(asyncio.CancelledError):
+                await task
+        getattr(session, "trigger_next_fire", {}).pop(trigger_id, None)
+
+        webhook_subs = getattr(session, "active_webhook_subs", {})
+        if sub_id := webhook_subs.pop(trigger_id, None):
+            with contextlib.suppress(Exception):
+                session.event_bus.unsubscribe(sub_id)
+
+        if tdef.trigger_type == "timer":
+            await _start_trigger_timer(session, trigger_id, tdef)
+        elif tdef.trigger_type == "webhook":
+            await _start_trigger_webhook(session, trigger_id, tdef)
+
+    if trigger_id in getattr(session, "active_trigger_ids", set()):
+        session_id = request.match_info["session_id"]
+        await _persist_active_triggers(session, session_id)
+
+    _save_trigger_to_agent(session, trigger_id, tdef)
+
+    # Emit SSE event so the frontend updates the graph and detail panel
+    bus = getattr(session, "event_bus", None)
+    if bus:
+        from framework.runtime.event_bus import AgentEvent, EventType
+
+        await bus.publish(
+            AgentEvent(
+                type=EventType.TRIGGER_UPDATED,
+                stream_id="queen",
+                data={
+                    "trigger_id": trigger_id,
+                    "task": tdef.task,
+                    "trigger_config": tdef.trigger_config,
+                    "trigger_type": tdef.trigger_type,
+                    "name": tdef.description or trigger_id,
+                    "entry_node": getattr(
+                        getattr(getattr(session, "runner", None), "graph", None),
+                        "entry_node",
+                        None,
+                    ),
+                },
+            )
+        )
+
     return web.json_response(
         {
-            "entry_points": [
-                {
-                    "id": ep.id,
-                    "name": ep.name,
-                    "entry_node": ep.entry_node,
-                    "trigger_type": ep.trigger_type,
-                    "trigger_config": ep.trigger_config,
-                    **(
-                        {"next_fire_in": nf}
-                        if rt and (nf := rt.get_timer_next_fire_in(ep.id)) is not None
-                        else {}
-                    ),
-                }
-                for ep in eps
-            ]
+            "trigger_id": trigger_id,
+            "task": tdef.task,
+            "trigger_config": tdef.trigger_config,
         }
     )
 
@@ -397,23 +587,28 @@ async def handle_list_worker_sessions(request: web.Request) -> web.Response:
     """List worker sessions on disk."""
     session, err = resolve_session(request)
     if err:
-        return err
-
-    if not session.worker_path:
-        return web.json_response({"sessions": []})
-
-    sess_dir = sessions_dir(session)
+        # Fall back to cold session lookup from disk
+        sid = request.match_info["session_id"]
+        sess_dir = cold_sessions_dir(sid)
+        if sess_dir is None:
+            return err
+    else:
+        if not session.worker_path:
+            return web.json_response({"sessions": []})
+        sess_dir = sessions_dir(session)
     if not sess_dir.exists():
         return web.json_response({"sessions": []})
 
     sessions = []
     for d in sorted(sess_dir.iterdir(), reverse=True):
-        if not d.is_dir() or not d.name.startswith("session_"):
+        if not d.is_dir():
+            continue
+        state_path = d / "state.json"
+        if not d.name.startswith("session_") and not state_path.exists():
             continue
 
         entry: dict = {"session_id": d.name}
 
-        state_path = d / "state.json"
         if state_path.exists():
             try:
                 state = json.loads(state_path.read_text(encoding="utf-8"))
@@ -564,48 +759,85 @@ async def handle_messages(request: web.Request) -> web.Response:
     """Get messages for a worker session."""
     session, err = resolve_session(request)
     if err:
-        return err
-
-    if not session.worker_path:
-        return web.json_response({"error": "No worker loaded"}, status=503)
+        # Fall back to cold session lookup from disk
+        sid = request.match_info["session_id"]
+        sess_dir = cold_sessions_dir(sid)
+        if sess_dir is None:
+            return err
+    else:
+        if not session.worker_path:
+            return web.json_response({"error": "No worker loaded"}, status=503)
+        sess_dir = sessions_dir(session)
 
     ws_id = request.match_info.get("ws_id") or request.match_info.get("session_id", "")
     ws_id = safe_path_segment(ws_id)
 
-    convs_dir = sessions_dir(session) / ws_id / "conversations"
+    convs_dir = sess_dir / ws_id / "conversations"
     if not convs_dir.exists():
         return web.json_response({"messages": []})
 
     filter_node = request.query.get("node_id")
     all_messages = []
 
-    for node_dir in convs_dir.iterdir():
-        if not node_dir.is_dir():
-            continue
-        if filter_node and node_dir.name != filter_node:
-            continue
-
-        parts_dir = node_dir / "parts"
+    def _collect_msg_parts(parts_dir: Path, node_id: str) -> None:
         if not parts_dir.exists():
-            continue
-
+            return
         for part_file in sorted(parts_dir.iterdir()):
             if part_file.suffix != ".json":
                 continue
             try:
                 part = json.loads(part_file.read_text(encoding="utf-8"))
-                part["_node_id"] = node_dir.name
+                part["_node_id"] = node_id
                 part.setdefault("created_at", part_file.stat().st_mtime)
                 all_messages.append(part)
             except (json.JSONDecodeError, OSError):
                 continue
 
+    # Flat layout: conversations/parts/*.json
+    if not filter_node:
+        _collect_msg_parts(convs_dir / "parts", "worker")
+
+    # Node-based layout: conversations/<node_id>/parts/*.json
+    for node_dir in convs_dir.iterdir():
+        if not node_dir.is_dir() or node_dir.name == "parts":
+            continue
+        if filter_node and node_dir.name != filter_node:
+            continue
+        _collect_msg_parts(node_dir / "parts", node_dir.name)
+
+    # Merge run lifecycle markers from runs.jsonl (for historical dividers)
+    runs_file = sess_dir / ws_id / "runs.jsonl"
+    if runs_file.exists():
+        try:
+            for line in runs_file.read_text(encoding="utf-8").splitlines():
+                line = line.strip()
+                if not line:
+                    continue
+                try:
+                    record = json.loads(line)
+                    all_messages.append(
+                        {
+                            "seq": -1,
+                            "role": "system",
+                            "content": "",
+                            "_node_id": "_run_marker",
+                            "is_run_marker": True,
+                            "run_id": record.get("run_id"),
+                            "run_event": record.get("event"),
+                            "created_at": record.get("created_at", 0),
+                        }
+                    )
+                except json.JSONDecodeError:
+                    continue
+        except OSError:
+            pass
+
     all_messages.sort(key=lambda m: m.get("created_at", m.get("seq", 0)))
 
     client_only = request.query.get("client_only", "").lower() in ("true", "1")
     if client_only:
         client_facing_nodes: set[str] = set()
-        if session.runner and hasattr(session.runner, "graph"):
+        if session and session.runner and hasattr(session.runner, "graph"):
             for node in session.runner.graph.nodes:
                 if node.client_facing:
                     client_facing_nodes.add(node.id)
@@ -614,63 +846,51 @@ async def handle_messages(request: web.Request) -> web.Response:
             all_messages = [
                 m
                 for m in all_messages
-                if not m.get("is_transition_marker")
-                and m["role"] != "tool"
-                and not (m["role"] == "assistant" and m.get("tool_calls"))
-                and (
-                    (m["role"] == "user" and m.get("is_client_input"))
-                    or (m["role"] == "assistant" and m.get("_node_id") in client_facing_nodes)
+                if m.get("is_run_marker")
+                or (
+                    not m.get("is_transition_marker")
+                    and m["role"] != "tool"
+                    and not (m["role"] == "assistant" and m.get("tool_calls"))
+                    and (
+                        (m["role"] == "user" and m.get("is_client_input"))
+                        or (m["role"] == "assistant" and m.get("_node_id") in client_facing_nodes)
+                    )
                 )
             ]
 
     return web.json_response({"messages": all_messages})
 
 
-async def handle_queen_messages(request: web.Request) -> web.Response:
-    """GET /api/sessions/{session_id}/queen-messages — get queen conversation.
+async def handle_session_events_history(request: web.Request) -> web.Response:
+    """GET /api/sessions/{session_id}/events/history — persisted eventbus log.
 
-    Reads directly from disk so it works for both live sessions and cold
-    (post-server-restart) sessions — no live session required.
+    Reads ``events.jsonl`` from the session directory on disk so it works for
+    both live sessions and cold (post-server-restart) sessions.  The frontend
+    replays these events through ``sseEventToChatMessage`` to fully reconstruct
+    the UI state on resume.
     """
     session_id = request.match_info["session_id"]
 
     queen_dir = Path.home() / ".hive" / "queen" / "session" / session_id
-    convs_dir = queen_dir / "conversations"
-    if not convs_dir.exists():
-        return web.json_response({"messages": [], "session_id": session_id})
+    events_path = queen_dir / "events.jsonl"
+    if not events_path.exists():
+        return web.json_response({"events": [], "session_id": session_id})
 
-    all_messages: list[dict] = []
-    for node_dir in convs_dir.iterdir():
-        if not node_dir.is_dir():
-            continue
-        parts_dir = node_dir / "parts"
-        if not parts_dir.exists():
-            continue
-        for part_file in sorted(parts_dir.iterdir()):
-            if part_file.suffix != ".json":
-                continue
-            try:
-                part = json.loads(part_file.read_text(encoding="utf-8"))
-                part["_node_id"] = node_dir.name
-                # Use file mtime as created_at so frontend can order
-                # queen and worker messages chronologically.
-                part.setdefault("created_at", part_file.stat().st_mtime)
-                all_messages.append(part)
-            except (json.JSONDecodeError, OSError):
-                continue
+    events: list[dict] = []
+    try:
+        with open(events_path, encoding="utf-8") as f:
+            for line in f:
+                line = line.strip()
+                if not line:
+                    continue
+                try:
+                    events.append(json.loads(line))
+                except json.JSONDecodeError:
+                    continue
+    except OSError:
+        return web.json_response({"events": [], "session_id": session_id})
 
-    all_messages.sort(key=lambda m: m.get("created_at", m.get("seq", 0)))
-
-    # Filter to client-facing messages only
-    all_messages = [
-        m
-        for m in all_messages
-        if not m.get("is_transition_marker")
-        and m["role"] != "tool"
-        and not (m["role"] == "assistant" and m.get("tool_calls"))
-    ]
-
-    return web.json_response({"messages": all_messages, "session_id": session_id})
+    return web.json_response({"events": events, "session_id": session_id})
 
 
 async def handle_session_history(request: web.Request) -> web.Response:
@@ -746,6 +966,7 @@ async def handle_discover(request: web.Request) -> web.Response:
                 "description": entry.description,
                 "category": entry.category,
                 "session_count": entry.session_count,
+                "run_count": entry.run_count,
                 "node_count": entry.node_count,
                 "tool_count": entry.tool_count,
                 "tags": entry.tags,
@@ -783,8 +1004,12 @@ def register_routes(app: web.Application) -> None:
     # Session info
     app.router.add_get("/api/sessions/{session_id}/stats", handle_session_stats)
     app.router.add_get("/api/sessions/{session_id}/entry-points", handle_session_entry_points)
+    app.router.add_patch(
+        "/api/sessions/{session_id}/triggers/{trigger_id}", handle_update_trigger_task
+    )
     app.router.add_get("/api/sessions/{session_id}/graphs", handle_session_graphs)
-    app.router.add_get("/api/sessions/{session_id}/queen-messages", handle_queen_messages)
+
+    app.router.add_get("/api/sessions/{session_id}/events/history", handle_session_events_history)
 
     # Worker session browsing (session-primary)
     app.router.add_get("/api/sessions/{session_id}/worker-sessions", handle_list_worker_sessions)

core/framework/server/session_manager.py

@@ -7,7 +7,6 @@ Architecture:
 - Session owns EventBus + LLM, shared with queen and worker
 - Queen is always present once a session starts
 - Worker is optional — loaded into an existing session
-- Judge is active only when a worker is loaded
 """
 
 import asyncio
@@ -15,11 +14,13 @@ import json
 import logging
 import time
 import uuid
-from dataclasses import dataclass
+from dataclasses import dataclass, field
 from datetime import datetime
 from pathlib import Path
 from typing import Any
 
+from framework.runtime.triggers import TriggerDefinition
+
 logger = logging.getLogger(__name__)
 
 
@@ -42,12 +43,25 @@ class Session:
     worker_info: Any | None = None  # AgentInfo
     # Queen phase state (building/staging/running)
     phase_state: Any = None  # QueenPhaseState
-    # Judge (active when worker is loaded)
-    judge_task: asyncio.Task | None = None
-    escalation_sub: str | None = None
+    # Worker handoff subscription
     worker_handoff_sub: str | None = None
     # Memory consolidation subscription (fires on CONTEXT_COMPACTED)
     memory_consolidation_sub: str | None = None
+    # Worker run digest subscription (fires on EXECUTION_COMPLETED / EXECUTION_FAILED)
+    worker_digest_sub: str | None = None
+    # Trigger definitions loaded from agent's triggers.json (available but inactive)
+    available_triggers: dict[str, TriggerDefinition] = field(default_factory=dict)
+    # Active trigger tracking (IDs currently firing + their asyncio tasks)
+    active_trigger_ids: set[str] = field(default_factory=set)
+    active_timer_tasks: dict[str, asyncio.Task] = field(default_factory=dict)
+    # Queen-owned webhook server (lazy singleton, created on first webhook trigger activation)
+    queen_webhook_server: Any = None
+    # EventBus subscription IDs for active webhook triggers (trigger_id -> sub_id)
+    active_webhook_subs: dict[str, str] = field(default_factory=dict)
+    # True after first successful worker execution (gates trigger delivery)
+    worker_configured: bool = False
+    # Monotonic timestamps for next trigger fire (mirrors AgentRuntime._timer_next_fire)
+    trigger_next_fire: dict[str, float] = field(default_factory=dict)
     # Session directory resumption:
     # When set, _start_queen writes queen conversations to this existing session's
     # directory instead of creating a new one.  This lets cold-restores accumulate
@@ -130,7 +144,9 @@ class SessionManager:
         to that existing session's directory instead of creating a new one.
         This preserves full conversation history across server restarts.
         """
-        session = await self._create_session_core(session_id=session_id, model=model)
+        # Reuse the original session ID when cold-restoring
+        resolved_session_id = queen_resume_from or session_id
+        session = await self._create_session_core(session_id=resolved_session_id, model=model)
         session.queen_resume_from = queen_resume_from
 
         # Start queen immediately (queen-only, no worker tools yet)
@@ -147,22 +163,53 @@ class SessionManager:
         self,
         agent_path: str | Path,
         agent_id: str | None = None,
+        session_id: str | None = None,
         model: str | None = None,
         initial_prompt: str | None = None,
         queen_resume_from: str | None = None,
     ) -> Session:
         """Create a session and load a worker in one step.
 
-        When ``queen_resume_from`` is set the queen writes conversation messages
-        to that existing session's directory instead of creating a new one.
+        When ``queen_resume_from`` is set the session reuses the original session
+        ID so the frontend sees a single continuous session.  The queen writes
+        conversation messages to that existing directory, preserving full history.
         """
         from framework.tools.queen_lifecycle_tools import build_worker_profile
 
         agent_path = Path(agent_path)
         resolved_worker_id = agent_id or agent_path.name
 
-        # Auto-generate session ID (not the agent name)
-        session = await self._create_session_core(model=model)
+        # When cold-restoring, check meta.json for the phase — if the agent
+        # was still being built we must NOT try to load the worker (the code
+        # is incomplete and will fail to import).
+        if queen_resume_from:
+            _resume_phase = None
+            _meta_path = (
+                Path.home() / ".hive" / "queen" / "session" / queen_resume_from / "meta.json"
+            )
+            if _meta_path.exists():
+                try:
+                    _meta = json.loads(_meta_path.read_text(encoding="utf-8"))
+                    _resume_phase = _meta.get("phase")
+                except (json.JSONDecodeError, OSError):
+                    pass
+            if _resume_phase in ("building", "planning"):
+                # Fall back to queen-only session — cold resume handler in
+                # _start_queen will set phase_state.agent_path and switch to
+                # the correct phase.
+                return await self.create_session(
+                    session_id=session_id,
+                    model=model,
+                    initial_prompt=initial_prompt,
+                    queen_resume_from=queen_resume_from,
+                )
+
+        # Reuse the original session ID when cold-restoring so the frontend
+        # sees one continuous session instead of a new one each time.
+        session = await self._create_session_core(
+            session_id=queen_resume_from,
+            model=model,
+        )
         session.queen_resume_from = queen_resume_from
         try:
             # Load worker FIRST (before queen) so queen gets full tools
@@ -173,6 +220,9 @@ class SessionManager:
                 model=model,
             )
 
+            # Restore active triggers from persisted state (cold restore)
+            await self._restore_active_triggers(session, session.id)
+
             # Start queen with worker profile + lifecycle + monitoring tools
             worker_identity = (
                 build_worker_profile(session.worker_runtime, agent_path=agent_path)
@@ -184,7 +234,23 @@ class SessionManager:
             )
 
         except Exception:
-            # If anything fails, tear down the session
+            if queen_resume_from:
+                # Cold restore: worker load failed (e.g. incomplete code from a
+                # building session).  Fall back to queen-only so the user can
+                # continue the conversation and fix / rebuild the agent.
+                logger.warning(
+                    "Cold restore: worker load failed for '%s', falling back to queen-only",
+                    agent_path,
+                    exc_info=True,
+                )
+                await self.stop_session(session.id)
+                return await self.create_session(
+                    session_id=session_id,
+                    model=model,
+                    initial_prompt=initial_prompt,
+                    queen_resume_from=queen_resume_from,
+                )
+            # If anything fails (non-cold-restore), tear down the session
             await self.stop_session(session.id)
             raise
         return session
@@ -202,8 +268,8 @@ class SessionManager:
     ) -> None:
         """Load a worker agent into a session (core logic).
 
-        Sets up the runner, runtime, and session fields. Does NOT start the
-        judge or notify the queen — callers handle those steps.
+        Sets up the runner, runtime, and session fields. Does NOT notify
+        the queen — callers handle that step.
         """
         from framework.runner import AgentRunner
 
@@ -242,6 +308,25 @@ class SessionManager:
 
             runtime = runner._agent_runtime
 
+            # Load triggers from the agent's triggers.json definition file.
+            from framework.tools.queen_lifecycle_tools import _read_agent_triggers_json
+
+            for tdata in _read_agent_triggers_json(agent_path):
+                tid = tdata.get("id", "")
+                ttype = tdata.get("trigger_type", "")
+                if tid and ttype in ("timer", "webhook"):
+                    session.available_triggers[tid] = TriggerDefinition(
+                        id=tid,
+                        trigger_type=ttype,
+                        trigger_config=tdata.get("trigger_config", {}),
+                        description=tdata.get("name", tid),
+                        task=tdata.get("task", ""),
+                    )
+                    logger.info("Loaded trigger '%s' (%s) from triggers.json", tid, ttype)
+
+            if session.available_triggers:
+                await self._emit_trigger_events(session, "available", session.available_triggers)
+
             # Start runtime on event loop
             if runtime and not runtime.is_running:
                 await runtime.start()
@@ -258,6 +343,9 @@ class SessionManager:
             session.worker_runtime = runtime
             session.worker_info = info
 
+            # Subscribe to execution completion for per-run digest generation
+            self._subscribe_worker_digest(session)
+
             async with self._lock:
                 self._loading.discard(session.id)
 
@@ -360,6 +448,51 @@ class SessionManager:
                 return False
             return True
 
+    async def _restore_active_triggers(self, session: "Session", session_id: str) -> None:
+        """Restore previously active triggers from persisted session state.
+
+        Called after worker loading to restart any timer/webhook triggers
+        that were active before a server restart.
+        """
+        if not session.available_triggers or not session.worker_runtime:
+            return
+        try:
+            store = session.worker_runtime._session_store
+            state = await store.read_state(session_id)
+            if state and state.active_triggers:
+                from framework.tools.queen_lifecycle_tools import (
+                    _start_trigger_timer,
+                    _start_trigger_webhook,
+                )
+
+                saved_tasks = getattr(state, "trigger_tasks", {}) or {}
+                for tid in state.active_triggers:
+                    tdef = session.available_triggers.get(tid)
+                    if tdef:
+                        # Restore user-configured task override
+                        saved_task = saved_tasks.get(tid, "")
+                        if saved_task:
+                            tdef.task = saved_task
+                        tdef.active = True
+                        session.active_trigger_ids.add(tid)
+                        if tdef.trigger_type == "timer":
+                            await _start_trigger_timer(session, tid, tdef)
+                            logger.info("Restored trigger timer '%s'", tid)
+                        elif tdef.trigger_type == "webhook":
+                            await _start_trigger_webhook(session, tid, tdef)
+                            logger.info("Restored webhook trigger '%s'", tid)
+                    else:
+                        logger.warning(
+                            "Saved trigger '%s' not found in worker entry points, skipping",
+                            tid,
+                        )
+
+            # Restore worker_configured flag
+            if state and getattr(state, "worker_configured", False):
+                session.worker_configured = True
+        except Exception as e:
+            logger.warning("Failed to restore active triggers: %s", e)
+
     async def load_worker(
         self,
         session_id: str,
@@ -369,7 +502,7 @@ class SessionManager:
     ) -> Session:
         """Load a worker agent into an existing session (with running queen).
 
-        Starts the worker runtime, health judge, and notifies the queen.
+        Starts the worker runtime and notifies the queen.
         """
         agent_path = Path(agent_path)
 
@@ -385,11 +518,31 @@ class SessionManager:
         )
 
         # Notify queen about the loaded worker (skip for queen itself).
-        # Health judge disabled for simplicity.
         if agent_path.name != "queen" and session.worker_runtime:
-            # await self._start_judge(session, session.runner._storage_path)
             await self._notify_queen_worker_loaded(session)
 
+        # Update meta.json so cold-restore can discover this session by agent_path
+        storage_session_id = session.queen_resume_from or session.id
+        meta_path = Path.home() / ".hive" / "queen" / "session" / storage_session_id / "meta.json"
+        try:
+            _agent_name = (
+                session.worker_info.name
+                if session.worker_info
+                else str(agent_path.name).replace("_", " ").title()
+            )
+            existing_meta = {}
+            if meta_path.exists():
+                existing_meta = json.loads(meta_path.read_text(encoding="utf-8"))
+            existing_meta["agent_name"] = _agent_name
+            existing_meta["agent_path"] = (
+                str(session.worker_path) if session.worker_path else str(agent_path)
+            )
+            meta_path.write_text(json.dumps(existing_meta), encoding="utf-8")
+        except OSError:
+            pass
+
+        await self._restore_active_triggers(session, session_id)
+
         # Emit SSE event so the frontend can update UI
         await self._emit_worker_loaded(session)
 
@@ -403,9 +556,6 @@ class SessionManager:
         if session.worker_runtime is None:
             return False
 
-        # Stop judge + escalation
-        self._stop_judge(session)
-
         # Cleanup worker
         if session.runner:
             try:
@@ -413,6 +563,33 @@ class SessionManager:
             except Exception as e:
                 logger.error("Error cleaning up worker '%s': %s", session.worker_id, e)
 
+        # Cancel active trigger timers
+        for tid, task in session.active_timer_tasks.items():
+            task.cancel()
+            logger.info("Cancelled trigger timer '%s' on unload", tid)
+        session.active_timer_tasks.clear()
+
+        # Unsubscribe webhook handlers (server stays alive — queen-owned)
+        for sub_id in session.active_webhook_subs.values():
+            try:
+                session.event_bus.unsubscribe(sub_id)
+            except Exception:
+                pass
+        session.active_webhook_subs.clear()
+        session.active_trigger_ids.clear()
+
+        # Clean up triggers
+        if session.available_triggers:
+            await self._emit_trigger_events(session, "removed", session.available_triggers)
+            session.available_triggers.clear()
+
+        if session.worker_digest_sub is not None:
+            try:
+                session.event_bus.unsubscribe(session.worker_digest_sub)
+            except Exception:
+                pass
+            session.worker_digest_sub = None
+
         worker_id = session.worker_id
         session.worker_id = None
         session.worker_path = None
@@ -443,8 +620,6 @@ class SessionManager:
         _storage_id = getattr(session, "queen_resume_from", None) or session_id
         _session_dir = Path.home() / ".hive" / "queen" / "session" / _storage_id
 
-        # Stop judge
-        self._stop_judge(session)
         if session.worker_handoff_sub is not None:
             try:
                 session.event_bus.unsubscribe(session.worker_handoff_sub)
@@ -452,6 +627,13 @@ class SessionManager:
                 pass
             session.worker_handoff_sub = None
 
+        if session.worker_digest_sub is not None:
+            try:
+                session.event_bus.unsubscribe(session.worker_digest_sub)
+            except Exception:
+                pass
+            session.worker_digest_sub = None
+
         # Stop queen and memory consolidation subscription
         if session.memory_consolidation_sub is not None:
             try:
@@ -464,6 +646,25 @@ class SessionManager:
             session.queen_task = None
         session.queen_executor = None
 
+        # Cancel active trigger timers
+        for task in session.active_timer_tasks.values():
+            task.cancel()
+        session.active_timer_tasks.clear()
+
+        # Unsubscribe webhook handlers and stop queen webhook server
+        for sub_id in session.active_webhook_subs.values():
+            try:
+                session.event_bus.unsubscribe(sub_id)
+            except Exception:
+                pass
+        session.active_webhook_subs.clear()
+        if session.queen_webhook_server is not None:
+            try:
+                await session.queen_webhook_server.stop()
+            except Exception:
+                logger.error("Error stopping queen webhook server", exc_info=True)
+            session.queen_webhook_server = None
+
         # Cleanup worker
         if session.runner:
             try:
@@ -482,6 +683,9 @@ class SessionManager:
                 name=f"queen-memory-consolidation-{session_id}",
             )
 
+        # Close per-session event log
+        session.event_bus.close_session_log()
+
         logger.info("Session '%s' stopped", session_id)
         return True
 
@@ -491,7 +695,7 @@ class SessionManager:
 
     async def _handle_worker_handoff(self, session: Session, executor: Any, event: Any) -> None:
         """Route worker escalation events into the queen conversation."""
-        if event.stream_id in ("queen", "judge"):
+        if event.stream_id == "queen":
             return
 
         reason = str(event.data.get("reason", "")).strip()
@@ -514,6 +718,134 @@ class SessionManager:
         else:
             logger.warning("Worker handoff received but queen node not ready")
 
+    def _subscribe_worker_digest(self, session: Session) -> None:
+        """Subscribe to worker events to write per-run digests.
+
+        Three triggers:
+        - NODE_LOOP_ITERATION: write a mid-run snapshot, throttled to at most
+          once every _DIGEST_COOLDOWN seconds per execution.
+        - TOOL_CALL_COMPLETED for delegate_to_sub_agent: same throttled snapshot.
+          Orchestrator nodes often run all subagent calls in a single LLM turn,
+          so NODE_LOOP_ITERATION only fires once at the end.  Subagent
+          completions provide intermediate checkpoints.
+        - EXECUTION_COMPLETED / EXECUTION_FAILED: always write the final digest,
+          bypassing the cooldown.
+        """
+        import time as _time
+
+        from framework.runtime.event_bus import EventType as _ET
+
+        _DIGEST_COOLDOWN = 300.0  # seconds between mid-run snapshots
+
+        if session.worker_digest_sub is not None:
+            try:
+                session.event_bus.unsubscribe(session.worker_digest_sub)
+            except Exception:
+                pass
+            session.worker_digest_sub = None
+
+        agent_name = session.worker_path.name if session.worker_path else None
+        if not agent_name:
+            return
+
+        _agent_name = agent_name
+        _llm = session.llm
+        _bus = session.event_bus
+        # per-execution_id monotonic timestamp of last mid-run digest
+        _last_digest: dict[str, float] = {}
+
+        def _resolve_run_id(exec_id: str) -> str | None:
+            """Look up the run_id for a given execution_id via EXECUTION_STARTED history."""
+            for e in _bus.get_history(event_type=_ET.EXECUTION_STARTED, limit=200):
+                if e.execution_id == exec_id and getattr(e, "run_id", None):
+                    return e.run_id
+            return None
+
+        async def _inject_digest_to_queen(run_id: str) -> None:
+            """Read the written digest and push it into the queen's conversation."""
+            from framework.agents.worker_memory import digest_path
+
+            try:
+                content = digest_path(_agent_name, run_id).read_text(encoding="utf-8").strip()
+            except OSError:
+                return
+            if not content:
+                return
+            executor = session.queen_executor
+            if executor is None:
+                return
+            node = executor.node_registry.get("queen")
+            if node is None or not hasattr(node, "inject_event"):
+                return
+            await node.inject_event(f"[WORKER_DIGEST]\n{content}")
+
+        async def _consolidate_and_notify(run_id: str, outcome_event: Any) -> None:
+            """Write the digest then push it to the queen."""
+            from framework.agents.worker_memory import consolidate_worker_run
+
+            await consolidate_worker_run(_agent_name, run_id, outcome_event, _bus, _llm)
+            await _inject_digest_to_queen(run_id)
+
+        async def _on_worker_event(event: Any) -> None:
+            if event.stream_id == "queen":
+                return
+
+            exec_id = event.execution_id
+
+            if event.type == _ET.EXECUTION_STARTED:
+                # New run on this execution_id — reset cooldown so the first
+                # iteration always produces a mid-run snapshot.
+                if exec_id:
+                    _last_digest.pop(exec_id, None)
+
+            elif event.type in (
+                _ET.EXECUTION_COMPLETED,
+                _ET.EXECUTION_FAILED,
+                _ET.EXECUTION_PAUSED,
+            ):
+                # Final digest — always fire, ignore cooldown.
+                # EXECUTION_PAUSED covers cancellation (queen re-triggering the
+                # worker cancels the previous execution, emitting paused).
+                run_id = getattr(event, "run_id", None) or _resolve_run_id(exec_id)
+                if run_id:
+                    asyncio.create_task(
+                        _consolidate_and_notify(run_id, event),
+                        name=f"worker-digest-final-{run_id}",
+                    )
+
+            elif event.type in (_ET.NODE_LOOP_ITERATION, _ET.TOOL_CALL_COMPLETED):
+                # Mid-run snapshot — respect 300 s cooldown per execution.
+                # TOOL_CALL_COMPLETED is only interesting for subagent calls;
+                # regular tool completions are too frequent and too cheap.
+                if event.type == _ET.TOOL_CALL_COMPLETED:
+                    tool_name = (event.data or {}).get("tool_name", "")
+                    if tool_name != "delegate_to_sub_agent":
+                        return
+                if not exec_id:
+                    return
+                now = _time.monotonic()
+                if now - _last_digest.get(exec_id, 0.0) < _DIGEST_COOLDOWN:
+                    return
+                run_id = _resolve_run_id(exec_id)
+                if run_id:
+                    _last_digest[exec_id] = now
+                    asyncio.create_task(
+                        _consolidate_and_notify(run_id, None),
+                        name=f"worker-digest-{run_id}",
+                    )
+
+        session.worker_digest_sub = session.event_bus.subscribe(
+            event_types=[
+                _ET.EXECUTION_STARTED,
+                _ET.NODE_LOOP_ITERATION,
+                _ET.TOOL_CALL_COMPLETED,
+                _ET.EXECUTION_COMPLETED,
+                _ET.EXECUTION_FAILED,
+                _ET.EXECUTION_PAUSED,
+            ],
+            handler=_on_worker_event,
+        )
+
     def _subscribe_worker_handoffs(self, session: Session, executor: Any) -> None:
         """Subscribe queen to worker/subagent escalation handoff events."""
         from framework.runtime.event_bus import EventType as _ET
@@ -567,19 +899,57 @@ class SessionManager:
                     else None
                 )
             )
-            _meta_path.write_text(
-                json.dumps(
-                    {
-                        "agent_name": _agent_name,
-                        "agent_path": str(session.worker_path) if session.worker_path else None,
-                        "created_at": time.time(),
-                    }
-                ),
-                encoding="utf-8",
-            )
+            # Merge into existing meta.json to preserve fields written by
+            # _update_meta_json (e.g. phase, agent_path set during building).
+            _existing_meta: dict = {}
+            if _meta_path.exists():
+                try:
+                    _existing_meta = json.loads(_meta_path.read_text(encoding="utf-8"))
+                except (json.JSONDecodeError, OSError):
+                    pass
+            _new_meta: dict = {"created_at": time.time()}
+            if _agent_name is not None:
+                _new_meta["agent_name"] = _agent_name
+            if session.worker_path is not None:
+                _new_meta["agent_path"] = str(session.worker_path)
+            _existing_meta.update(_new_meta)
+            _meta_path.write_text(json.dumps(_existing_meta), encoding="utf-8")
         except OSError:
             pass
 
+        # Enable per-session event persistence so that all eventbus events
+        # survive server restarts and can be replayed on cold-session resume.
+        # Scan the existing event log to find the max iteration ever written,
+        # then use max+1 as offset so resumed sessions produce monotonically
+        # increasing iteration values — preventing frontend message ID collisions.
+        iteration_offset = 0
+        events_path = queen_dir / "events.jsonl"
+        try:
+            if events_path.exists():
+                max_iter = -1
+                with open(events_path, encoding="utf-8") as f:
+                    for line in f:
+                        line = line.strip()
+                        if not line:
+                            continue
+                        try:
+                            evt = json.loads(line)
+                            it = evt.get("data", {}).get("iteration")
+                            if isinstance(it, int) and it > max_iter:
+                                max_iter = it
+                        except (json.JSONDecodeError, TypeError):
+                            continue
+                if max_iter >= 0:
+                    iteration_offset = max_iter + 1
+                    logger.info(
+                        "Session '%s' resuming with iteration_offset=%d (from events.jsonl max)",
+                        session.id,
+                        iteration_offset,
+                    )
+        except OSError:
+            pass
+        session.event_bus.set_session_log(events_path, iteration_offset=iteration_offset)
+
         session.queen_task = await create_queen(
             session=session,
             session_manager=self,
@@ -588,6 +958,38 @@ class SessionManager:
             initial_prompt=initial_prompt,
         )
 
+        # Auto-load worker on cold restore — the queen's conversation expects
+        # the agent to be loaded, but the new session has no worker.
+        if session.queen_resume_from and not session.worker_runtime:
+            meta_path = queen_dir / "meta.json"
+            if meta_path.exists():
+                try:
+                    _meta = json.loads(meta_path.read_text(encoding="utf-8"))
+                    _agent_path = _meta.get("agent_path")
+                    _phase = _meta.get("phase")
+
+                    if _agent_path and Path(_agent_path).exists():
+                        if _phase in ("staging", "running", None):
+                            # Agent fully built — load worker and resume
+                            await self.load_worker(session.id, _agent_path)
+                            if session.phase_state:
+                                await session.phase_state.switch_to_staging(source="auto")
+                            # Emit flowchart overlay so frontend can display it
+                            await self._emit_flowchart_on_restore(session, _agent_path)
+                            logger.info("Cold restore: auto-loaded worker from %s", _agent_path)
+                        elif _phase == "building":
+                            # Agent folder exists but incomplete — resume building
+                            if session.phase_state:
+                                session.phase_state.agent_path = _agent_path
+                                await session.phase_state.switch_to_building(source="auto")
+                            logger.info("Cold restore: resumed BUILDING phase for %s", _agent_path)
+                        elif _phase == "planning":
+                            if session.phase_state:
+                                session.phase_state.agent_path = _agent_path
+                            logger.info("Cold restore: PLANNING phase for %s", _agent_path)
+                except Exception:
+                    logger.warning("Cold restore: failed to auto-load worker", exc_info=True)
+
         # Memory consolidation — triggered by context compaction events.
         # Compaction is a natural signal that "enough has happened to be worth remembering".
         _consolidation_llm = session.llm
@@ -607,116 +1009,6 @@ class SessionManager:
             handler=_on_compaction,
         )
 
-    # ------------------------------------------------------------------
-    # Judge startup / teardown
-    # ------------------------------------------------------------------
-
-    async def _start_judge(
-        self,
-        session: Session,
-        worker_storage_path: str | Path,
-    ) -> None:
-        """Start the health judge for a session's worker."""
-        from framework.graph.executor import GraphExecutor
-        from framework.monitoring import judge_goal, judge_graph
-        from framework.runner.tool_registry import ToolRegistry
-        from framework.runtime.core import Runtime
-        from framework.runtime.event_bus import EventType as _ET
-        from framework.tools.worker_monitoring_tools import register_worker_monitoring_tools
-
-        worker_storage_path = Path(worker_storage_path)
-
-        try:
-            # Monitoring tools
-            monitoring_registry = ToolRegistry()
-            register_worker_monitoring_tools(
-                monitoring_registry,
-                session.event_bus,
-                worker_storage_path,
-                worker_graph_id=session.worker_runtime._graph_id,
-            )
-
-            hive_home = Path.home() / ".hive"
-            judge_dir = hive_home / "judge" / "session" / session.id
-            judge_dir.mkdir(parents=True, exist_ok=True)
-
-            judge_runtime = Runtime(hive_home / "judge")
-            monitoring_tools = list(monitoring_registry.get_tools().values())
-            monitoring_executor = monitoring_registry.get_executor()
-
-            async def _judge_loop():
-                interval = 300  # 5 minutes between checks
-                # Wait before the first check — let the worker actually do something
-                await asyncio.sleep(interval)
-                while True:
-                    try:
-                        executor = GraphExecutor(
-                            runtime=judge_runtime,
-                            llm=session.llm,
-                            tools=monitoring_tools,
-                            tool_executor=monitoring_executor,
-                            event_bus=session.event_bus,
-                            stream_id="judge",
-                            storage_path=judge_dir,
-                            loop_config=judge_graph.loop_config,
-                        )
-                        await executor.execute(
-                            graph=judge_graph,
-                            goal=judge_goal,
-                            input_data={
-                                "event": {"source": "timer", "reason": "scheduled"},
-                            },
-                            session_state={"resume_session_id": session.id},
-                        )
-                    except Exception:
-                        logger.error("Health judge tick failed", exc_info=True)
-                    await asyncio.sleep(interval)
-
-            session.judge_task = asyncio.create_task(_judge_loop())
-
-            # Escalation: judge → queen
-            async def _on_escalation(event):
-                ticket = event.data.get("ticket", {})
-                executor = session.queen_executor
-                if executor is None:
-                    logger.warning("Escalation received but queen executor is None")
-                    return
-                node = executor.node_registry.get("queen")
-                if node is not None and hasattr(node, "inject_event"):
-                    msg = "[ESCALATION TICKET from Health Judge]\n" + json.dumps(
-                        ticket, indent=2, ensure_ascii=False
-                    )
-                    await node.inject_event(msg)
-                else:
-                    logger.warning("Escalation received but queen node not ready")
-
-            session.escalation_sub = session.event_bus.subscribe(
-                event_types=[_ET.WORKER_ESCALATION_TICKET],
-                handler=_on_escalation,
-            )
-
-            logger.info("Judge started for session '%s'", session.id)
-
-        except Exception as e:
-            logger.error(
-                "Failed to start judge for session '%s': %s",
-                session.id,
-                e,
-                exc_info=True,
-            )
-
-    def _stop_judge(self, session: Session) -> None:
-        """Cancel judge task and unsubscribe escalation events."""
-        if session.judge_task is not None:
-            session.judge_task.cancel()
-            session.judge_task = None
-        if session.escalation_sub is not None:
-            try:
-                session.event_bus.unsubscribe(session.escalation_sub)
-            except Exception:
-                pass
-            session.escalation_sub = None
-
     # ------------------------------------------------------------------
     # Queen notifications
     # ------------------------------------------------------------------
@@ -733,7 +1025,22 @@ class SessionManager:
             return
 
         profile = build_worker_profile(session.worker_runtime, agent_path=session.worker_path)
-        await node.inject_event(f"[SYSTEM] Worker loaded.{profile}")
+
+        # Append available trigger info so the queen knows what's schedulable
+        trigger_lines = ""
+        if session.available_triggers:
+            parts = []
+            for t in session.available_triggers.values():
+                cfg = t.trigger_config
+                detail = cfg.get("cron") or f"every {cfg.get('interval_minutes', '?')} min"
+                task_info = f' -> task: "{t.task}"' if t.task else " (no task configured)"
+                parts.append(f"  - {t.id} ({t.trigger_type}: {detail}){task_info}")
+            trigger_lines = (
+                "\n\nAvailable triggers (inactive — use set_trigger to activate):\n"
+                + "\n".join(parts)
+            )
+
+        await node.inject_event(f"[SYSTEM] Worker loaded.{profile}{trigger_lines}")
 
     async def _emit_worker_loaded(self, session: Session) -> None:
         """Publish a WORKER_LOADED event so the frontend can update."""
@@ -754,6 +1061,29 @@ class SessionManager:
             )
         )
 
+    async def _emit_flowchart_on_restore(self, session: Session, agent_path: str | Path) -> None:
+        """Emit FLOWCHART_MAP_UPDATED from persisted flowchart file on cold restore."""
+        from framework.runtime.event_bus import AgentEvent, EventType
+        from framework.tools.flowchart_utils import load_flowchart_file
+
+        original_draft, flowchart_map = load_flowchart_file(agent_path)
+        if original_draft is None:
+            return
+        # Cache in phase_state so the REST endpoint also returns it
+        if session.phase_state:
+            session.phase_state.original_draft_graph = original_draft
+            session.phase_state.flowchart_map = flowchart_map
+        await session.event_bus.publish(
+            AgentEvent(
+                type=EventType.FLOWCHART_MAP_UPDATED,
+                stream_id="queen",
+                data={
+                    "map": flowchart_map,
+                    "original_draft": original_draft,
+                },
+            )
+        )
+
     async def _notify_queen_worker_unloaded(self, session: Session) -> None:
         """Notify the queen that the worker has been unloaded."""
         executor = session.queen_executor
@@ -765,9 +1095,41 @@ class SessionManager:
 
         await node.inject_event(
             "[SYSTEM] Worker unloaded. You are now operating independently. "
-            "Handle all tasks directly using your coding tools."
+            "Design or build the agent to solve the user's problem "
+            "according to your current phase."
         )
 
+    async def _emit_trigger_events(
+        self,
+        session: Session,
+        kind: str,
+        triggers: dict[str, TriggerDefinition],
+    ) -> None:
+        """Emit TRIGGER_AVAILABLE or TRIGGER_REMOVED events for each trigger."""
+        from framework.runtime.event_bus import AgentEvent, EventType
+
+        event_type = (
+            EventType.TRIGGER_AVAILABLE if kind == "available" else EventType.TRIGGER_REMOVED
+        )
+        # Resolve graph entry node for trigger target
+        runner = getattr(session, "runner", None)
+        graph_entry = runner.graph.entry_node if runner else None
+
+        for t in triggers.values():
+            await session.event_bus.publish(
+                AgentEvent(
+                    type=event_type,
+                    stream_id="queen",
+                    data={
+                        "trigger_id": t.id,
+                        "trigger_type": t.trigger_type,
+                        "trigger_config": t.trigger_config,
+                        "name": t.description or t.id,
+                        **({"entry_node": graph_entry} if graph_entry else {}),
+                    },
+                )
+            )
+
     async def revive_queen(self, session: Session, initial_prompt: str | None = None) -> None:
         """Revive a dead queen executor on an existing session.
 
@@ -839,13 +1201,19 @@ class SessionManager:
         # Check whether any message part files are actually present
         has_messages = False
         try:
-            for node_dir in convs_dir.iterdir():
-                if not node_dir.is_dir():
-                    continue
-                parts_dir = node_dir / "parts"
-                if parts_dir.exists() and any(f.suffix == ".json" for f in parts_dir.iterdir()):
-                    has_messages = True
-                    break
+            # Flat layout: conversations/parts/*.json
+            flat_parts = convs_dir / "parts"
+            if flat_parts.exists() and any(f.suffix == ".json" for f in flat_parts.iterdir()):
+                has_messages = True
+            else:
+                # Node-based layout: conversations/<node_id>/parts/*.json
+                for node_dir in convs_dir.iterdir():
+                    if not node_dir.is_dir() or node_dir.name == "parts":
+                        continue
+                    parts_dir = node_dir / "parts"
+                    if parts_dir.exists() and any(f.suffix == ".json" for f in parts_dir.iterdir()):
+                        has_messages = True
+                        break
         except OSError:
             pass
 
@@ -922,21 +1290,27 @@ class SessionManager:
             if convs_dir.exists():
                 try:
                     all_parts: list[dict] = []
-                    for node_dir in convs_dir.iterdir():
-                        if not node_dir.is_dir():
-                            continue
-                        parts_dir = node_dir / "parts"
+
+                    def _collect_parts(parts_dir: Path, _dest: list[dict] = all_parts) -> None:
                         if not parts_dir.exists():
-                            continue
+                            return
                         for part_file in sorted(parts_dir.iterdir()):
                             if part_file.suffix != ".json":
                                 continue
                             try:
                                 part = json.loads(part_file.read_text(encoding="utf-8"))
                                 part.setdefault("created_at", part_file.stat().st_mtime)
-                                all_parts.append(part)
+                                _dest.append(part)
                             except (json.JSONDecodeError, OSError):
                                 continue
+
+                    # Flat layout: conversations/parts/*.json
+                    _collect_parts(convs_dir / "parts")
+                    # Node-based layout: conversations/<node_id>/parts/*.json
+                    for node_dir in convs_dir.iterdir():
+                        if not node_dir.is_dir() or node_dir.name == "parts":
+                            continue
+                        _collect_parts(node_dir / "parts")
                     # Filter to client-facing messages only
                     client_msgs = [
                         p

core/framework/server/tests/test_api.py

@@ -5,6 +5,7 @@ Uses aiohttp TestClient with mocked sessions to test all endpoints
 without requiring actual LLM calls or agent loading.
 """
 
+import asyncio
 import json
 from dataclasses import dataclass, field
 from pathlib import Path
@@ -13,9 +14,13 @@ from unittest.mock import AsyncMock, MagicMock
 import pytest
 from aiohttp.test_utils import TestClient, TestServer
 
+from framework.runtime.triggers import TriggerDefinition
 from framework.server.app import create_app
 from framework.server.session_manager import Session
 
+REPO_ROOT = Path(__file__).resolve().parents[4]
+EXAMPLE_AGENT_PATH = REPO_ROOT / "examples" / "templates" / "deep_research_agent"
+
 # ---------------------------------------------------------------------------
 # Mock helpers
 # ---------------------------------------------------------------------------
@@ -169,6 +174,7 @@ def _make_session(
     runner.intro_message = "Test intro"
 
     mock_event_bus = MagicMock()
+    mock_event_bus.publish = AsyncMock()
     mock_llm = MagicMock()
 
     queen_executor = _make_queen_executor() if with_queen else None
@@ -207,11 +213,8 @@ def tmp_agent_dir(tmp_path, monkeypatch):
     return tmp_path, agent_name, base
 
 
-@pytest.fixture
-def sample_session(tmp_agent_dir):
-    """Create a sample session with state.json, checkpoints, and conversations."""
-    tmp_path, agent_name, base = tmp_agent_dir
-    session_id = "session_20260220_120000_abc12345"
+def _write_sample_session(base: Path, session_id: str):
+    """Create a sample worker session on disk."""
     session_dir = base / "sessions" / session_id
 
     # state.json
@@ -292,6 +295,20 @@ def sample_session(tmp_agent_dir):
     return session_id, session_dir, state
 
 
+@pytest.fixture
+def sample_session(tmp_agent_dir):
+    """Create a sample session with state.json, checkpoints, and conversations."""
+    _tmp_path, _agent_name, base = tmp_agent_dir
+    return _write_sample_session(base, "session_20260220_120000_abc12345")
+
+
+@pytest.fixture
+def custom_id_session(tmp_agent_dir):
+    """Create a sample session that uses a custom non-session_* ID."""
+    _tmp_path, _agent_name, base = tmp_agent_dir
+    return _write_sample_session(base, "my-custom-session")
+
+
 def _make_app_with_session(session):
     """Create an aiohttp app with a pre-loaded session."""
     app = create_app()
@@ -347,6 +364,35 @@ class TestHealth:
 
 
 class TestSessionCRUD:
+    @pytest.mark.asyncio
+    async def test_create_session_with_worker_forwards_session_id(self):
+        app = create_app()
+        manager = app["manager"]
+        manager.create_session_with_worker = AsyncMock(
+            return_value=_make_session(agent_id="my-custom-session")
+        )
+
+        async with TestClient(TestServer(app)) as client:
+            resp = await client.post(
+                "/api/sessions",
+                json={
+                    "session_id": "my-custom-session",
+                    "agent_path": str(EXAMPLE_AGENT_PATH),
+                },
+            )
+            data = await resp.json()
+
+        assert resp.status == 201
+        assert data["session_id"] == "my-custom-session"
+        manager.create_session_with_worker.assert_awaited_once_with(
+            str(EXAMPLE_AGENT_PATH.resolve()),
+            agent_id=None,
+            session_id="my-custom-session",
+            model=None,
+            initial_prompt=None,
+            queen_resume_from=None,
+        )
+
     @pytest.mark.asyncio
     async def test_list_sessions_empty(self):
         app = create_app()
@@ -441,6 +487,70 @@ class TestSessionCRUD:
             data = await resp.json()
             assert "primary" in data["graphs"]
 
+    @pytest.mark.asyncio
+    async def test_update_trigger_task(self, tmp_path):
+        session = _make_session(tmp_dir=tmp_path)
+        session.available_triggers["daily"] = TriggerDefinition(
+            id="daily",
+            trigger_type="timer",
+            trigger_config={"cron": "0 5 * * *"},
+            task="Old task",
+        )
+        app = _make_app_with_session(session)
+        async with TestClient(TestServer(app)) as client:
+            resp = await client.patch(
+                "/api/sessions/test_agent/triggers/daily",
+                json={"task": "New task"},
+            )
+            assert resp.status == 200
+            data = await resp.json()
+            assert data["task"] == "New task"
+            assert data["trigger_config"]["cron"] == "0 5 * * *"
+            assert session.available_triggers["daily"].task == "New task"
+
+    @pytest.mark.asyncio
+    async def test_update_trigger_cron_restarts_active_timer(self, tmp_path):
+        session = _make_session(tmp_dir=tmp_path)
+        session.available_triggers["daily"] = TriggerDefinition(
+            id="daily",
+            trigger_type="timer",
+            trigger_config={"cron": "0 5 * * *"},
+            task="Run task",
+            active=True,
+        )
+        session.active_trigger_ids.add("daily")
+        session.active_timer_tasks["daily"] = asyncio.create_task(asyncio.sleep(60))
+        app = _make_app_with_session(session)
+        async with TestClient(TestServer(app)) as client:
+            resp = await client.patch(
+                "/api/sessions/test_agent/triggers/daily",
+                json={"trigger_config": {"cron": "0 6 * * *"}},
+            )
+            assert resp.status == 200
+            data = await resp.json()
+            assert data["trigger_config"]["cron"] == "0 6 * * *"
+            assert "daily" in session.active_timer_tasks
+            assert session.active_timer_tasks["daily"] is not None
+            assert session.available_triggers["daily"].trigger_config["cron"] == "0 6 * * *"
+            session.active_timer_tasks["daily"].cancel()
+
+    @pytest.mark.asyncio
+    async def test_update_trigger_cron_rejects_invalid_expression(self, tmp_path):
+        session = _make_session(tmp_dir=tmp_path)
+        session.available_triggers["daily"] = TriggerDefinition(
+            id="daily",
+            trigger_type="timer",
+            trigger_config={"cron": "0 5 * * *"},
+            task="Run task",
+        )
+        app = _make_app_with_session(session)
+        async with TestClient(TestServer(app)) as client:
+            resp = await client.patch(
+                "/api/sessions/test_agent/triggers/daily",
+                json={"trigger_config": {"cron": "not a cron"}},
+            )
+            assert resp.status == 400
+
 
 class TestExecution:
     @pytest.mark.asyncio
@@ -767,6 +877,22 @@ class TestWorkerSessions:
             assert data["sessions"][0]["status"] == "paused"
             assert data["sessions"][0]["steps"] == 5
 
+    @pytest.mark.asyncio
+    async def test_list_sessions_includes_custom_id(self, custom_id_session, tmp_agent_dir):
+        session_id, session_dir, state = custom_id_session
+        tmp_path, agent_name, base = tmp_agent_dir
+
+        session = _make_session(tmp_dir=tmp_path / ".hive" / "agents" / agent_name)
+        app = _make_app_with_session(session)
+
+        async with TestClient(TestServer(app)) as client:
+            resp = await client.get("/api/sessions/test_agent/worker-sessions")
+            assert resp.status == 200
+            data = await resp.json()
+            assert len(data["sessions"]) == 1
+            assert data["sessions"][0]["session_id"] == session_id
+            assert data["sessions"][0]["status"] == "paused"
+
     @pytest.mark.asyncio
     async def test_list_sessions_empty(self, tmp_agent_dir):
         tmp_path, agent_name, base = tmp_agent_dir
@@ -1284,6 +1410,28 @@ class TestLogs:
             assert len(data["logs"]) >= 1
             assert data["logs"][0]["run_id"] == session_id
 
+    @pytest.mark.asyncio
+    async def test_logs_list_summaries_with_custom_id(self, custom_id_session, tmp_agent_dir):
+        session_id, session_dir, state = custom_id_session
+        tmp_path, agent_name, base = tmp_agent_dir
+
+        from framework.runtime.runtime_log_store import RuntimeLogStore
+
+        log_store = RuntimeLogStore(base)
+        session = _make_session(
+            tmp_dir=tmp_path / ".hive" / "agents" / agent_name,
+            log_store=log_store,
+        )
+        app = _make_app_with_session(session)
+
+        async with TestClient(TestServer(app)) as client:
+            resp = await client.get("/api/sessions/test_agent/logs")
+            assert resp.status == 200
+            data = await resp.json()
+            assert "logs" in data
+            assert len(data["logs"]) >= 1
+            assert data["logs"][0]["run_id"] == session_id
+
     @pytest.mark.asyncio
     async def test_logs_session_summary(self, sample_session, tmp_agent_dir):
         session_id, session_dir, state = sample_session

core/framework/skills/__init__.py

@@ -0,0 +1,26 @@
+"""Hive Agent Skills — discovery, parsing, and injection of SKILL.md packages.
+
+Implements the open Agent Skills standard (agentskills.io) for portable
+skill discovery and activation, plus built-in default skills for runtime
+operational discipline.
+"""
+
+from framework.skills.catalog import SkillCatalog
+from framework.skills.config import DefaultSkillConfig, SkillsConfig
+from framework.skills.defaults import DefaultSkillManager
+from framework.skills.discovery import DiscoveryConfig, SkillDiscovery
+from framework.skills.manager import SkillsManager, SkillsManagerConfig
+from framework.skills.parser import ParsedSkill, parse_skill_md
+
+__all__ = [
+    "DefaultSkillConfig",
+    "DefaultSkillManager",
+    "DiscoveryConfig",
+    "ParsedSkill",
+    "SkillCatalog",
+    "SkillDiscovery",
+    "SkillsConfig",
+    "SkillsManager",
+    "SkillsManagerConfig",
+    "parse_skill_md",
+]

core/framework/skills/_default_skills/batch-ledger/SKILL.md

@@ -0,0 +1,24 @@
+---
+name: hive.batch-ledger
+description: Track per-item status when processing collections to prevent skipped or duplicated items.
+metadata:
+  author: hive
+  type: default-skill
+---
+
+## Operational Protocol: Batch Progress Ledger
+
+When processing a collection of items, maintain a batch ledger in `_batch_ledger`.
+
+Initialize when you identify the batch:
+- `_batch_total`: total item count
+- `_batch_ledger`: JSON with per-item status
+
+Per-item statuses: pending → in_progress → completed|failed|skipped
+
+- Set `in_progress` BEFORE processing
+- Set final status AFTER processing with 1-line result_summary
+- Include error reason for failed/skipped items
+- Update aggregate counts after each item
+- NEVER remove items from the ledger
+- If resuming, skip items already marked completed

core/framework/skills/_default_skills/context-preservation/SKILL.md

@@ -0,0 +1,22 @@
+---
+name: hive.context-preservation
+description: Proactively preserve critical information before automatic context pruning destroys it.
+metadata:
+  author: hive
+  type: default-skill
+---
+
+## Operational Protocol: Context Preservation
+
+You operate under a finite context window. Important information WILL be pruned.
+
+Save-As-You-Go: After any tool call producing information you'll need later,
+immediately extract key data into `_working_notes` or `_preserved_data`.
+Do NOT rely on referring back to old tool results.
+
+What to extract: URLs and key snippets (not full pages), relevant API fields
+(not raw JSON), specific lines/values (not entire files), analysis results
+(not raw data).
+
+Before transitioning to the next phase/node, write a handoff summary to
+`_handoff_context` with everything the next phase needs to know.

core/framework/skills/_default_skills/error-recovery/SKILL.md

@@ -0,0 +1,18 @@
+---
+name: hive.error-recovery
+description: Follow a structured recovery protocol when tool calls fail instead of blindly retrying or giving up.
+metadata:
+  author: hive
+  type: default-skill
+---
+
+## Operational Protocol: Error Recovery
+
+When a tool call fails:
+
+1. Diagnose — record error in notes, classify as transient or structural
+2. Decide — transient: retry once. Structural fixable: fix and retry.
+   Structural unfixable: record as failed, move to next item.
+   Blocking all progress: record escalation note.
+3. Adapt — if same tool failed 3+ times, stop using it and find alternative.
+   Update plan in notes. Never silently drop the failed item.

core/framework/skills/_default_skills/note-taking/SKILL.md

@@ -0,0 +1,27 @@
+---
+name: hive.note-taking
+description: Maintain structured working notes throughout execution to prevent information loss during context pruning.
+metadata:
+  author: hive
+  type: default-skill
+---
+
+## Operational Protocol: Structured Note-Taking
+
+Maintain structured working notes in shared memory key `_working_notes`.
+Update at these checkpoints:
+
+- After completing each discrete subtask or batch item
+- After receiving new information that changes your plan
+- Before any tool call that will produce substantial output
+
+Structure:
+
+### Objective — restate the goal
+### Current Plan — numbered steps, mark completed with ✓
+### Key Decisions — decisions made and WHY
+### Working Data — intermediate results, extracted values
+### Open Questions — uncertainties to verify
+### Blockers — anything preventing progress
+
+Update incrementally — do not rewrite from scratch each time.

core/framework/skills/_default_skills/quality-monitor/SKILL.md

@@ -0,0 +1,20 @@
+---
+name: hive.quality-monitor
+description: Periodically self-assess output quality to catch degradation before the judge does.
+metadata:
+  author: hive
+  type: default-skill
+---
+
+## Operational Protocol: Quality Self-Assessment
+
+Every 5 iterations, self-assess:
+
+1. On-task? Still working toward the stated objective?
+2. Thorough? Cutting corners compared to earlier?
+3. Non-repetitive? Producing new value or rehashing?
+4. Consistent? Latest output contradict earlier decisions?
+5. Complete? Tracking all items, or silently dropped some?
+
+If degrading: write assessment to `_quality_log`, re-read `_working_notes`,
+change approach explicitly. If acceptable: brief note in `_quality_log`.

core/framework/skills/_default_skills/task-decomposition/SKILL.md

@@ -0,0 +1,17 @@
+---
+name: hive.task-decomposition
+description: Decompose complex tasks into explicit subtasks before diving in.
+metadata:
+  author: hive
+  type: default-skill
+---
+
+## Operational Protocol: Task Decomposition
+
+Before starting a complex task:
+
+1. Decompose — break into numbered subtasks in `_working_notes` Current Plan
+2. Estimate — relative effort per subtask (small/medium/large)
+3. Execute — work through in order, mark ✓ when complete
+4. Budget — if running low on iterations, prioritize by impact
+5. Verify — before declaring done, every subtask must be ✓, skipped (with reason), or blocked

core/framework/skills/catalog.py

@@ -0,0 +1,107 @@
+"""Skill catalog — in-memory index with system prompt generation.
+
+Builds the XML catalog injected into the system prompt for model-driven
+skill activation per the Agent Skills standard.
+"""
+
+from __future__ import annotations
+
+import logging
+from xml.sax.saxutils import escape
+
+from framework.skills.parser import ParsedSkill
+
+logger = logging.getLogger(__name__)
+
+_BEHAVIORAL_INSTRUCTION = (
+    "The following skills provide specialized instructions for specific tasks.\n"
+    "When a task matches a skill's description, read the SKILL.md at the listed\n"
+    "location to load the full instructions before proceeding.\n"
+    "When a skill references relative paths, resolve them against the skill's\n"
+    "directory (the parent of SKILL.md) and use absolute paths in tool calls."
+)
+
+
+class SkillCatalog:
+    """In-memory catalog of discovered skills."""
+
+    def __init__(self, skills: list[ParsedSkill] | None = None):
+        self._skills: dict[str, ParsedSkill] = {}
+        self._activated: set[str] = set()
+        if skills:
+            for skill in skills:
+                self.add(skill)
+
+    def add(self, skill: ParsedSkill) -> None:
+        """Add a skill to the catalog."""
+        self._skills[skill.name] = skill
+
+    def get(self, name: str) -> ParsedSkill | None:
+        """Look up a skill by name."""
+        return self._skills.get(name)
+
+    def mark_activated(self, name: str) -> None:
+        """Mark a skill as activated in the current session."""
+        self._activated.add(name)
+
+    def is_activated(self, name: str) -> bool:
+        """Check if a skill has been activated."""
+        return name in self._activated
+
+    @property
+    def skill_count(self) -> int:
+        return len(self._skills)
+
+    @property
+    def allowlisted_dirs(self) -> list[str]:
+        """All skill base directories for file access allowlisting."""
+        return [skill.base_dir for skill in self._skills.values()]
+
+    def to_prompt(self) -> str:
+        """Generate the catalog prompt for system prompt injection.
+
+        Returns empty string if no community/user skills are discovered
+        (default skills are handled separately by DefaultSkillManager).
+        """
+        # Filter out framework-scope skills (default skills) — they're
+        # injected via the protocols prompt, not the catalog
+        community_skills = [s for s in self._skills.values() if s.source_scope != "framework"]
+
+        if not community_skills:
+            return ""
+
+        lines = ["<available_skills>"]
+        for skill in sorted(community_skills, key=lambda s: s.name):
+            lines.append("  <skill>")
+            lines.append(f"    <name>{escape(skill.name)}</name>")
+            lines.append(f"    <description>{escape(skill.description)}</description>")
+            lines.append(f"    <location>{escape(skill.location)}</location>")
+            lines.append("  </skill>")
+        lines.append("</available_skills>")
+
+        xml_block = "\n".join(lines)
+        return f"{_BEHAVIORAL_INSTRUCTION}\n\n{xml_block}"
+
+    def build_pre_activated_prompt(self, skill_names: list[str]) -> str:
+        """Build prompt content for pre-activated skills.
+
+        Pre-activated skills get their full SKILL.md body loaded into
+        the system prompt at startup (tier 2), bypassing model-driven
+        activation.
+
+        Returns empty string if no skills match.
+        """
+        parts: list[str] = []
+
+        for name in skill_names:
+            skill = self.get(name)
+            if skill is None:
+                logger.warning("Pre-activated skill '%s' not found in catalog", name)
+                continue
+            if self.is_activated(name):
+                continue  # Already activated, skip duplicate
+
+            self.mark_activated(name)
+            parts.append(f"--- Pre-Activated Skill: {skill.name} ---\n{skill.body}")
+
+        return "\n\n".join(parts)

core/framework/skills/config.py

@@ -0,0 +1,100 @@
+"""Skill configuration dataclasses.
+
+Handles agent-level skill configuration from module-level variables
+(``default_skills`` and ``skills``).
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+from typing import Any
+
+
+@dataclass
+class DefaultSkillConfig:
+    """Configuration for a single default skill."""
+
+    enabled: bool = True
+    overrides: dict[str, Any] = field(default_factory=dict)
+
+    @classmethod
+    def from_dict(cls, data: dict[str, Any]) -> DefaultSkillConfig:
+        enabled = data.get("enabled", True)
+        overrides = {k: v for k, v in data.items() if k != "enabled"}
+        return cls(enabled=enabled, overrides=overrides)
+
+
+@dataclass
+class SkillsConfig:
+    """Agent-level skill configuration.
+
+    Built from module-level variables in agent.py::
+
+        # Pre-activated community skills
+        skills = ["deep-research", "code-review"]
+
+        # Default skill configuration
+        default_skills = {
+            "hive.note-taking": {"enabled": True},
+            "hive.batch-ledger": {"enabled": True, "checkpoint_every_n": 10},
+            "hive.quality-monitor": {"enabled": False},
+        }
+    """
+
+    # Per-default-skill config, keyed by skill name (e.g. "hive.note-taking")
+    default_skills: dict[str, DefaultSkillConfig] = field(default_factory=dict)
+
+    # Pre-activated community skills (by name)
+    skills: list[str] = field(default_factory=list)
+
+    # Master switch: disable all default skills at once
+    all_defaults_disabled: bool = False
+
+    def is_default_enabled(self, skill_name: str) -> bool:
+        """Check if a specific default skill is enabled."""
+        if self.all_defaults_disabled:
+            return False
+        config = self.default_skills.get(skill_name)
+        if config is None:
+            return True  # enabled by default
+        return config.enabled
+
+    def get_default_overrides(self, skill_name: str) -> dict[str, Any]:
+        """Get skill-specific configuration overrides."""
+        config = self.default_skills.get(skill_name)
+        if config is None:
+            return {}
+        return config.overrides
+
+    @classmethod
+    def from_agent_vars(
+        cls,
+        default_skills: dict[str, Any] | None = None,
+        skills: list[str] | None = None,
+    ) -> SkillsConfig:
+        """Build config from agent module-level variables.
+
+        Args:
+            default_skills: Dict from agent module, e.g.
+                ``{"hive.note-taking": {"enabled": True}}``
+            skills: List of pre-activated skill names from agent module
+        """
+        all_disabled = False
+        parsed_defaults: dict[str, DefaultSkillConfig] = {}
+
+        if default_skills:
+            for name, config_dict in default_skills.items():
+                if name == "_all":
+                    if isinstance(config_dict, dict) and not config_dict.get("enabled", True):
+                        all_disabled = True
+                    continue
+                if isinstance(config_dict, dict):
+                    parsed_defaults[name] = DefaultSkillConfig.from_dict(config_dict)
+                elif isinstance(config_dict, bool):
+                    parsed_defaults[name] = DefaultSkillConfig(enabled=config_dict)
+
+        return cls(
+            default_skills=parsed_defaults,
+            skills=list(skills or []),
+            all_defaults_disabled=all_disabled,
+        )

core/framework/skills/defaults.py

@@ -0,0 +1,151 @@
+"""DefaultSkillManager — load, configure, and inject built-in default skills.
+
+Default skills are SKILL.md packages shipped with the framework that provide
+runtime operational protocols (note-taking, batch tracking, error recovery, etc.).
+"""
+
+from __future__ import annotations
+
+import logging
+from pathlib import Path
+
+from framework.skills.config import SkillsConfig
+from framework.skills.parser import ParsedSkill, parse_skill_md
+
+logger = logging.getLogger(__name__)
+
+# Default skills directory relative to this module
+_DEFAULT_SKILLS_DIR = Path(__file__).parent / "_default_skills"
+
+# Ordered list of default skills (name → directory)
+SKILL_REGISTRY: dict[str, str] = {
+    "hive.note-taking": "note-taking",
+    "hive.batch-ledger": "batch-ledger",
+    "hive.context-preservation": "context-preservation",
+    "hive.quality-monitor": "quality-monitor",
+    "hive.error-recovery": "error-recovery",
+    "hive.task-decomposition": "task-decomposition",
+}
+
+# All shared memory keys used by default skills (for permission auto-inclusion)
+SHARED_MEMORY_KEYS: list[str] = [
+    # note-taking
+    "_working_notes",
+    "_notes_updated_at",
+    # batch-ledger
+    "_batch_ledger",
+    "_batch_total",
+    "_batch_completed",
+    "_batch_failed",
+    # context-preservation
+    "_handoff_context",
+    "_preserved_data",
+    # quality-monitor
+    "_quality_log",
+    "_quality_degradation_count",
+    # error-recovery
+    "_error_log",
+    "_failed_tools",
+    "_escalation_needed",
+    # task-decomposition
+    "_subtasks",
+    "_iteration_budget_remaining",
+]
+
+
+class DefaultSkillManager:
+    """Manages loading, configuration, and prompt generation for default skills."""
+
+    def __init__(self, config: SkillsConfig | None = None):
+        self._config = config or SkillsConfig()
+        self._skills: dict[str, ParsedSkill] = {}
+        self._loaded = False
+
+    def load(self) -> None:
+        """Load all enabled default skill SKILL.md files."""
+        if self._loaded:
+            return
+
+        for skill_name, dir_name in SKILL_REGISTRY.items():
+            if not self._config.is_default_enabled(skill_name):
+                logger.info("Default skill '%s' disabled by config", skill_name)
+                continue
+
+            skill_path = _DEFAULT_SKILLS_DIR / dir_name / "SKILL.md"
+            if not skill_path.is_file():
+                logger.error("Default skill SKILL.md not found: %s", skill_path)
+                continue
+
+            parsed = parse_skill_md(skill_path, source_scope="framework")
+            if parsed is None:
+                logger.error("Failed to parse default skill: %s", skill_path)
+                continue
+
+            self._skills[skill_name] = parsed
+
+        self._loaded = True
+
+    def build_protocols_prompt(self) -> str:
+        """Build the combined operational protocols section.
+
+        Extracts protocol sections from all enabled default skills and
+        combines them into a single ``## Operational Protocols`` block
+        for system prompt injection.
+
+        Returns empty string if all defaults are disabled.
+        """
+        if not self._skills:
+            return ""
+
+        parts: list[str] = ["## Operational Protocols\n"]
+
+        for skill_name in SKILL_REGISTRY:
+            skill = self._skills.get(skill_name)
+            if skill is None:
+                continue
+            # Use the full body — each SKILL.md contains exactly one protocol section
+            parts.append(skill.body)
+
+        if len(parts) <= 1:
+            return ""
+
+        combined = "\n\n".join(parts)
+
+        # Token budget warning (approximate: 1 token ≈ 4 chars)
+        approx_tokens = len(combined) // 4
+        if approx_tokens > 2000:
+            logger.warning(
+                "Default skill protocols exceed 2000 token budget "
+                "(~%d tokens, %d chars). Consider trimming.",
+                approx_tokens,
+                len(combined),
+            )
+
+        return combined
+
+    def log_active_skills(self) -> None:
+        """Log which default skills are active and their configuration."""
+        if not self._skills:
+            logger.info("Default skills: all disabled")
+            return
+
+        active = []
+        for skill_name in SKILL_REGISTRY:
+            if skill_name in self._skills:
+                overrides = self._config.get_default_overrides(skill_name)
+                if overrides:
+                    active.append(f"{skill_name} ({overrides})")
+                else:
+                    active.append(skill_name)
+
+        logger.info("Default skills active: %s", ", ".join(active))
+
+    @property
+    def active_skill_names(self) -> list[str]:
+        """Names of all currently active default skills."""
+        return list(self._skills.keys())
+
+    @property
+    def active_skills(self) -> dict[str, ParsedSkill]:
+        """All active default skills keyed by name."""
+        return dict(self._skills)

core/framework/skills/discovery.py

@@ -0,0 +1,183 @@
+"""Skill discovery — scan standard directories for SKILL.md files.
+
+Implements the Agent Skills standard discovery paths plus Hive-specific
+locations. Resolves name collisions deterministically.
+"""
+
+from __future__ import annotations
+
+import logging
+from dataclasses import dataclass
+from pathlib import Path
+
+from framework.skills.parser import ParsedSkill, parse_skill_md
+
+logger = logging.getLogger(__name__)
+
+# Directories to skip during scanning
+_SKIP_DIRS = frozenset(
+    {
+        ".git",
+        "node_modules",
+        "__pycache__",
+        ".venv",
+        "venv",
+        ".mypy_cache",
+        ".pytest_cache",
+        ".ruff_cache",
+    }
+)
+
+# Scope priority (higher = takes precedence)
+_SCOPE_PRIORITY = {
+    "framework": 0,
+    "user": 1,
+    "project": 2,
+}
+
+# Within the same scope, Hive-specific paths override cross-client paths.
+# We encode this by scanning cross-client first, then Hive-specific (later wins).
+
+
+@dataclass
+class DiscoveryConfig:
+    """Configuration for skill discovery."""
+
+    project_root: Path | None = None
+    skip_user_scope: bool = False
+    skip_framework_scope: bool = False
+    max_depth: int = 4
+    max_dirs: int = 2000
+
+
+class SkillDiscovery:
+    """Scans standard directories for SKILL.md files and resolves collisions."""
+
+    def __init__(self, config: DiscoveryConfig | None = None):
+        self._config = config or DiscoveryConfig()
+
+    def discover(self) -> list[ParsedSkill]:
+        """Scan all scopes and return deduplicated skill list.
+
+        Scanning order (lowest to highest precedence):
+        1. Framework defaults
+        2. User cross-client (~/.agents/skills/)
+        3. User Hive-specific (~/.hive/skills/)
+        4. Project cross-client (<project>/.agents/skills/)
+        5. Project Hive-specific (<project>/.hive/skills/)
+
+        Later entries override earlier ones on name collision.
+        """
+        all_skills: list[ParsedSkill] = []
+
+        # Framework scope (lowest precedence)
+        if not self._config.skip_framework_scope:
+            framework_dir = Path(__file__).parent / "_default_skills"
+            if framework_dir.is_dir():
+                all_skills.extend(self._scan_scope(framework_dir, "framework"))
+
+        # User scope
+        if not self._config.skip_user_scope:
+            home = Path.home()
+
+            # Cross-client (lower precedence within user scope)
+            user_agents = home / ".agents" / "skills"
+            if user_agents.is_dir():
+                all_skills.extend(self._scan_scope(user_agents, "user"))
+
+            # Hive-specific (higher precedence within user scope)
+            user_hive = home / ".hive" / "skills"
+            if user_hive.is_dir():
+                all_skills.extend(self._scan_scope(user_hive, "user"))
+
+        # Project scope (highest precedence)
+        if self._config.project_root:
+            root = self._config.project_root
+
+            # Cross-client
+            project_agents = root / ".agents" / "skills"
+            if project_agents.is_dir():
+                all_skills.extend(self._scan_scope(project_agents, "project"))
+
+            # Hive-specific
+            project_hive = root / ".hive" / "skills"
+            if project_hive.is_dir():
+                all_skills.extend(self._scan_scope(project_hive, "project"))
+
+        resolved = self._resolve_collisions(all_skills)
+
+        logger.info(
+            "Skill discovery: found %d skills (%d after dedup) across all scopes",
+            len(all_skills),
+            len(resolved),
+        )
+        return resolved
+
+    def _scan_scope(self, root: Path, scope: str) -> list[ParsedSkill]:
+        """Scan a single directory for skill directories containing SKILL.md."""
+        skills: list[ParsedSkill] = []
+        dirs_scanned = 0
+
+        for skill_md in self._find_skill_files(root, depth=0):
+            if dirs_scanned >= self._config.max_dirs:
+                logger.warning(
+                    "Hit max directory limit (%d) scanning %s",
+                    self._config.max_dirs,
+                    root,
+                )
+                break
+
+            parsed = parse_skill_md(skill_md, source_scope=scope)
+            if parsed is not None:
+                skills.append(parsed)
+            dirs_scanned += 1
+
+        return skills
+
+    def _find_skill_files(self, directory: Path, depth: int) -> list[Path]:
+        """Recursively find SKILL.md files up to max_depth."""
+        if depth > self._config.max_depth:
+            return []
+
+        results: list[Path] = []
+
+        try:
+            entries = sorted(directory.iterdir())
+        except OSError:
+            return []
+
+        for entry in entries:
+            if not entry.is_dir():
+                continue
+            if entry.name in _SKIP_DIRS:
+                continue
+
+            skill_md = entry / "SKILL.md"
+            if skill_md.is_file():
+                results.append(skill_md)
+            else:
+                # Recurse into subdirectories
+                results.extend(self._find_skill_files(entry, depth + 1))
+
+        return results
+
+    def _resolve_collisions(self, skills: list[ParsedSkill]) -> list[ParsedSkill]:
+        """Resolve name collisions deterministically.
+
+        Later entries in the list override earlier ones (because we scan
+        from lowest to highest precedence). On collision, log a warning.
+        """
+        seen: dict[str, ParsedSkill] = {}
+
+        for skill in skills:
+            if skill.name in seen:
+                existing = seen[skill.name]
+                logger.warning(
+                    "Skill name collision: '%s' from %s overrides %s",
+                    skill.name,
+                    skill.location,
+                    existing.location,
+                )
+            seen[skill.name] = skill
+
+        return list(seen.values())

core/framework/skills/manager.py

@@ -0,0 +1,165 @@
+"""Unified skill lifecycle manager.
+
+``SkillsManager`` is the single facade that owns skill discovery, loading,
+and prompt renderation.  The runtime creates one at startup and downstream
+layers read the cached prompt strings.
+
+Typical usage — **config-driven** (runner passes configuration)::
+
+    config = SkillsManagerConfig(
+        skills_config=SkillsConfig.from_agent_vars(...),
+        project_root=agent_path,
+    )
+    mgr = SkillsManager(config)
+    mgr.load()
+    print(mgr.protocols_prompt)       # default skill protocols
+    print(mgr.skills_catalog_prompt)  # community skills XML
+
+Typical usage — **bare** (exported agents, SDK users)::
+
+    mgr = SkillsManager()   # default config
+    mgr.load()               # loads all 6 default skills, no community discovery
+"""
+
+from __future__ import annotations
+
+import logging
+from dataclasses import dataclass, field
+from pathlib import Path
+
+from framework.skills.config import SkillsConfig
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class SkillsManagerConfig:
+    """Everything the runtime needs to configure skills.
+
+    Attributes:
+        skills_config: Per-skill enable/disable and overrides.
+        project_root: Agent directory for community skill discovery.
+            When ``None``, community discovery is skipped.
+        skip_community_discovery: Explicitly skip community scanning
+            even when ``project_root`` is set.
+    """
+
+    skills_config: SkillsConfig = field(default_factory=SkillsConfig)
+    project_root: Path | None = None
+    skip_community_discovery: bool = False
+
+
+class SkillsManager:
+    """Unified skill lifecycle: discovery → loading → prompt renderation.
+
+    The runtime creates one instance during init and owns it for the
+    lifetime of the process.  Downstream layers (``ExecutionStream``,
+    ``GraphExecutor``, ``NodeContext``, ``EventLoopNode``) receive the
+    cached prompt strings via property accessors.
+    """
+
+    def __init__(self, config: SkillsManagerConfig | None = None) -> None:
+        self._config = config or SkillsManagerConfig()
+        self._loaded = False
+        self._catalog_prompt: str = ""
+        self._protocols_prompt: str = ""
+
+    # ------------------------------------------------------------------
+    # Factory for backwards-compat bridge
+    # ------------------------------------------------------------------
+
+    @classmethod
+    def from_precomputed(
+        cls,
+        skills_catalog_prompt: str = "",
+        protocols_prompt: str = "",
+    ) -> SkillsManager:
+        """Wrap pre-rendered prompt strings (legacy callers).
+
+        Returns a manager that skips discovery/loading and just returns
+        the provided strings.  Used by the deprecation bridge in
+        ``AgentRuntime`` when callers pass raw prompt strings.
+        """
+        mgr = cls.__new__(cls)
+        mgr._config = SkillsManagerConfig()
+        mgr._loaded = True  # skip load()
+        mgr._catalog_prompt = skills_catalog_prompt
+        mgr._protocols_prompt = protocols_prompt
+        return mgr
+
+    # ------------------------------------------------------------------
+    # Lifecycle
+    # ------------------------------------------------------------------
+
+    def load(self) -> None:
+        """Discover, load, and cache skill prompts.  Idempotent."""
+        if self._loaded:
+            return
+        self._loaded = True
+
+        try:
+            self._do_load()
+        except Exception:
+            logger.warning("Skill system init failed (non-fatal)", exc_info=True)
+
+    def _do_load(self) -> None:
+        """Internal load — may raise; caller catches."""
+        from framework.skills.catalog import SkillCatalog
+        from framework.skills.defaults import DefaultSkillManager
+        from framework.skills.discovery import DiscoveryConfig, SkillDiscovery
+
+        skills_config = self._config.skills_config
+
+        # 1. Community skill discovery (when project_root is available)
+        catalog_prompt = ""
+        if self._config.project_root is not None and not self._config.skip_community_discovery:
+            discovery = SkillDiscovery(DiscoveryConfig(project_root=self._config.project_root))
+            discovered = discovery.discover()
+            catalog = SkillCatalog(discovered)
+            catalog_prompt = catalog.to_prompt()
+
+            # Pre-activated community skills
+            if skills_config.skills:
+                pre_activated = catalog.build_pre_activated_prompt(skills_config.skills)
+                if pre_activated:
+                    if catalog_prompt:
+                        catalog_prompt = f"{catalog_prompt}\n\n{pre_activated}"
+                    else:
+                        catalog_prompt = pre_activated
+
+        # 2. Default skills (always loaded unless explicitly disabled)
+        default_mgr = DefaultSkillManager(config=skills_config)
+        default_mgr.load()
+        default_mgr.log_active_skills()
+        protocols_prompt = default_mgr.build_protocols_prompt()
+
+        # 3. Cache
+        self._catalog_prompt = catalog_prompt
+        self._protocols_prompt = protocols_prompt
+
+        if protocols_prompt:
+            logger.info(
+                "Skill system ready: protocols=%d chars, catalog=%d chars",
+                len(protocols_prompt),
+                len(catalog_prompt),
+            )
+        else:
+            logger.warning("Skill system produced empty protocols_prompt")
+
+    # ------------------------------------------------------------------
+    # Prompt accessors (consumed by downstream layers)
+    # ------------------------------------------------------------------
+
+    @property
+    def skills_catalog_prompt(self) -> str:
+        """Community skills XML catalog for system prompt injection."""
+        return self._catalog_prompt
+
+    @property
+    def protocols_prompt(self) -> str:
+        """Default skill operational protocols for system prompt injection."""
+        return self._protocols_prompt
+
+    @property
+    def is_loaded(self) -> bool:
+        return self._loaded

core/framework/skills/parser.py

@@ -0,0 +1,158 @@
+"""SKILL.md parser — extracts YAML frontmatter and markdown body.
+
+Parses SKILL.md files per the Agent Skills standard (agentskills.io/specification).
+Lenient validation: warns on non-critical issues, skips only on missing description
+or completely unparseable YAML.
+"""
+
+from __future__ import annotations
+
+import logging
+import re
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Any
+
+logger = logging.getLogger(__name__)
+
+# Maximum name length before a warning is logged
+_MAX_NAME_LENGTH = 64
+
+
+@dataclass
+class ParsedSkill:
+    """In-memory representation of a parsed SKILL.md file."""
+
+    name: str
+    description: str
+    location: str  # absolute path to SKILL.md
+    base_dir: str  # parent directory of SKILL.md
+    source_scope: str  # "project", "user", or "framework"
+    body: str  # markdown body after closing ---
+
+    # Optional frontmatter fields
+    license: str | None = None
+    compatibility: list[str] | None = None
+    metadata: dict[str, Any] | None = None
+    allowed_tools: list[str] | None = None
+
+
+def _try_fix_yaml(raw: str) -> str:
+    """Attempt to fix common YAML issues (unquoted colon values).
+
+    Some SKILL.md files written for other clients may contain unquoted
+    values with colons, e.g. ``description: Use for: research tasks``.
+    This wraps such values in quotes as a best-effort fixup.
+    """
+    lines = raw.split("\n")
+    fixed = []
+    for line in lines:
+        # Match "key: value" where value contains an unquoted colon
+        m = re.match(r"^(\s*\w[\w-]*:\s*)(.+)$", line)
+        if m:
+            key_part, value_part = m.group(1), m.group(2)
+            # If value contains a colon and isn't already quoted
+            if ":" in value_part and not (value_part.startswith('"') or value_part.startswith("'")):
+                value_part = f'"{value_part}"'
+            fixed.append(f"{key_part}{value_part}")
+        else:
+            fixed.append(line)
+    return "\n".join(fixed)
+
+
+def parse_skill_md(path: Path, source_scope: str = "project") -> ParsedSkill | None:
+    """Parse a SKILL.md file into a ParsedSkill record.
+
+    Args:
+        path: Absolute path to the SKILL.md file.
+        source_scope: One of "project", "user", or "framework".
+
+    Returns:
+        ParsedSkill on success, None if the file is unparseable or
+        missing required fields (description).
+    """
+    try:
+        content = path.read_text(encoding="utf-8")
+    except OSError as exc:
+        logger.error("Failed to read %s: %s", path, exc)
+        return None
+
+    if not content.strip():
+        logger.error("Empty SKILL.md: %s", path)
+        return None
+
+    # Split on --- delimiters (first two occurrences)
+    parts = content.split("---", 2)
+    if len(parts) < 3:
+        logger.error("SKILL.md missing YAML frontmatter delimiters (---): %s", path)
+        return None
+
+    # parts[0] is content before first --- (should be empty or whitespace)
+    # parts[1] is the YAML frontmatter
+    # parts[2] is the markdown body
+    raw_yaml = parts[1].strip()
+    body = parts[2].strip()
+
+    if not raw_yaml:
+        logger.error("Empty YAML frontmatter in %s", path)
+        return None
+
+    # Parse YAML
+    import yaml
+
+    frontmatter: dict[str, Any] | None = None
+    try:
+        frontmatter = yaml.safe_load(raw_yaml)
+    except yaml.YAMLError:
+        # Fallback: try fixing unquoted colon values
+        try:
+            fixed = _try_fix_yaml(raw_yaml)
+            frontmatter = yaml.safe_load(fixed)
+            logger.warning("Fixed YAML parse issues in %s (unquoted colons)", path)
+        except yaml.YAMLError as exc:
+            logger.error("Unparseable YAML in %s: %s", path, exc)
+            return None
+
+    if not isinstance(frontmatter, dict):
+        logger.error("YAML frontmatter is not a mapping in %s", path)
+        return None
+
+    # Required: description
+    description = frontmatter.get("description")
+    if not description or not str(description).strip():
+        logger.error("Missing or empty 'description' in %s — skipping skill", path)
+        return None
+
+    # Required: name (fallback to parent directory name)
+    name = frontmatter.get("name")
+    parent_dir_name = path.parent.name
+    if not name or not str(name).strip():
+        name = parent_dir_name
+        logger.warning("Missing 'name' in %s — using directory name '%s'", path, name)
+    else:
+        name = str(name).strip()
+
+    # Lenient warnings
+    if len(name) > _MAX_NAME_LENGTH:
+        logger.warning("Skill name exceeds %d chars in %s: '%s'", _MAX_NAME_LENGTH, path, name)
+
+    if name != parent_dir_name and not name.endswith(f".{parent_dir_name}"):
+        logger.warning(
+            "Skill name '%s' doesn't match parent directory '%s' in %s",
+            name,
+            parent_dir_name,
+            path,
+        )
+
+    return ParsedSkill(
+        name=name,
+        description=str(description).strip(),
+        location=str(path.resolve()),
+        base_dir=str(path.parent.resolve()),
+        source_scope=source_scope,
+        body=body,
+        license=frontmatter.get("license"),
+        compatibility=frontmatter.get("compatibility"),
+        metadata=frontmatter.get("metadata"),
+        allowed_tools=frontmatter.get("allowed-tools"),
+    )

core/framework/testing/llm_judge.py

@@ -40,18 +40,31 @@ class LLMJudge:
 
     def _get_fallback_provider(self) -> LLMProvider | None:
         """
-        Auto-detects available API keys and returns the appropriate provider.
-        Priority: OpenAI -> Anthropic.
+        Auto-detects available API keys and returns an appropriate provider.
+        Uses LiteLLM for OpenAI (framework has no framework.llm.openai module).
+        Priority:
+        1. OpenAI-compatible models via LiteLLM (OPENAI_API_KEY)
+        2. Anthropic via AnthropicProvider (ANTHROPIC_API_KEY)
         """
+        # OpenAI: use LiteLLM (the framework's standard multi-provider integration)
         if os.environ.get("OPENAI_API_KEY"):
-            from framework.llm.openai import OpenAIProvider
+            try:
+                from framework.llm.litellm import LiteLLMProvider
 
-            return OpenAIProvider(model="gpt-4o-mini")
+                return LiteLLMProvider(model="gpt-4o-mini")
+            except ImportError:
+                # LiteLLM is optional; fall through to Anthropic/None
+                pass
 
+        # Anthropic via dedicated provider (wraps LiteLLM internally)
         if os.environ.get("ANTHROPIC_API_KEY"):
-            from framework.llm.anthropic import AnthropicProvider
+            try:
+                from framework.llm.anthropic import AnthropicProvider
 
-            return AnthropicProvider(model="claude-3-haiku-20240307")
+                return AnthropicProvider(model="claude-haiku-4-5-20251001")
+            except Exception:
+                # If AnthropicProvider cannot be constructed, treat as no fallback
+                return None
 
         return None
 
@@ -77,11 +90,16 @@ SUMMARY TO EVALUATE:
 Respond with JSON: {{"passes": true/false, "explanation": "..."}}"""
 
         try:
+            # Compute fallback provider once so we do not create multiple instances
+            fallback_provider = self._get_fallback_provider()
+
             # 1. Use injected provider
             if self._provider:
                 active_provider = self._provider
-            # 2. Check if _get_client was MOCKED (legacy tests) or use Agnostic Fallback
-            elif hasattr(self._get_client, "return_value") or not self._get_fallback_provider():
+            # 2. Legacy path: anthropic client mocked in tests takes precedence,
+            #    or no fallback provider is available.
+            elif hasattr(self._get_client, "return_value") or fallback_provider is None:
+                # Use legacy Anthropic client (e.g. when tests mock _get_client, or no env keys set)
                 client = self._get_client()
                 response = client.messages.create(
                     model="claude-haiku-4-5-20251001",
@@ -90,7 +108,8 @@ Respond with JSON: {{"passes": true/false, "explanation": "..."}}"""
                 )
                 return self._parse_json_result(response.content[0].text.strip())
             else:
-                active_provider = self._get_fallback_provider()
+                # Use env-based fallback (LiteLLM or AnthropicProvider)
+                active_provider = fallback_provider
 
             response = active_provider.complete(
                 messages=[{"role": "user", "content": prompt}],

core/framework/tools/flowchart_utils.py

@@ -0,0 +1,374 @@
+"""Flowchart utilities for generating and persisting flowchart.json files.
+
+Extracted from queen_lifecycle_tools so that non-Queen code paths
+(e.g., AgentRunner.load) can generate flowcharts for legacy agents
+that lack a flowchart.json.
+"""
+
+from __future__ import annotations
+
+import json
+import logging
+from pathlib import Path
+from typing import Any
+
+logger = logging.getLogger(__name__)
+
+FLOWCHART_FILENAME = "flowchart.json"
+
+# ── Flowchart type catalogue (9 types) ───────────────────────────────────────
+FLOWCHART_TYPES = {
+    "start": {"shape": "stadium", "color": "#8aad3f"},  # spring pollen
+    "terminal": {"shape": "stadium", "color": "#b5453a"},  # propolis red
+    "process": {"shape": "rectangle", "color": "#b5a575"},  # warm wheat
+    "decision": {"shape": "diamond", "color": "#d89d26"},  # royal honey
+    "io": {"shape": "parallelogram", "color": "#d06818"},  # burnt orange
+    "document": {"shape": "document", "color": "#c4b830"},  # goldenrod
+    "database": {"shape": "cylinder", "color": "#508878"},  # sage teal
+    "subprocess": {"shape": "subroutine", "color": "#887a48"},  # propolis gold
+    "browser": {"shape": "hexagon", "color": "#cc8850"},  # honey copper
+}
+
+# Backward-compat remap: old type names → canonical type
+FLOWCHART_REMAP: dict[str, str] = {
+    "delay": "process",
+    "manual_operation": "process",
+    "preparation": "process",
+    "merge": "process",
+    "alternate_process": "process",
+    "connector": "process",
+    "offpage_connector": "process",
+    "extract": "process",
+    "sort": "process",
+    "collate": "process",
+    "summing_junction": "process",
+    "or": "process",
+    "comment": "process",
+    "display": "io",
+    "manual_input": "io",
+    "multi_document": "document",
+    "stored_data": "database",
+    "internal_storage": "database",
+}
+
+
+# ── File persistence ─────────────────────────────────────────────────────────
+
+
+def save_flowchart_file(
+    agent_path: Path | str | None,
+    original_draft: dict,
+    flowchart_map: dict[str, list[str]] | None,
+) -> None:
+    """Persist the flowchart to the agent's folder."""
+    if agent_path is None:
+        return
+    p = Path(agent_path)
+    if not p.is_dir():
+        return
+    try:
+        target = p / FLOWCHART_FILENAME
+        target.write_text(
+            json.dumps(
+                {"original_draft": original_draft, "flowchart_map": flowchart_map},
+                indent=2,
+            ),
+            encoding="utf-8",
+        )
+        logger.debug("Flowchart saved to %s", target)
+    except Exception:
+        logger.warning("Failed to save flowchart to %s", p, exc_info=True)
+
+
+def load_flowchart_file(
+    agent_path: Path | str | None,
+) -> tuple[dict | None, dict[str, list[str]] | None]:
+    """Load flowchart from the agent's folder. Returns (original_draft, flowchart_map)."""
+    if agent_path is None:
+        return None, None
+    target = Path(agent_path) / FLOWCHART_FILENAME
+    if not target.is_file():
+        return None, None
+    try:
+        data = json.loads(target.read_text(encoding="utf-8"))
+        return data.get("original_draft"), data.get("flowchart_map")
+    except Exception:
+        logger.warning("Failed to load flowchart from %s", target, exc_info=True)
+        return None, None
+
+
+# ── Node classification ──────────────────────────────────────────────────────
+
+
+def classify_flowchart_node(
+    node: dict,
+    index: int,
+    total: int,
+    edges: list[dict],
+    terminal_ids: set[str],
+) -> str:
+    """Auto-detect the ISO 5807 flowchart type for a draft node.
+
+    Priority: explicit override > structural detection > heuristic > default.
+    """
+    # Explicit override from the queen
+    explicit = node.get("flowchart_type", "").strip()
+    if explicit and explicit in FLOWCHART_TYPES:
+        return explicit
+    if explicit and explicit in FLOWCHART_REMAP:
+        return FLOWCHART_REMAP[explicit]
+
+    node_id = node["id"]
+    node_type = node.get("node_type", "event_loop")
+    node_tools = set(node.get("tools") or [])
+    desc = (node.get("description") or "").lower()
+
+    # GCU / browser automation nodes → hexagon
+    if node_type == "gcu":
+        return "browser"
+
+    # Entry node (first node or no incoming edges) → start terminator
+    incoming = {e["target"] for e in edges}
+    if index == 0 or (node_id not in incoming and index == 0):
+        return "start"
+
+    # Terminal node → end terminator
+    if node_id in terminal_ids:
+        return "terminal"
+
+    # Decision node: has outgoing edges with branching conditions → diamond
+    outgoing = [e for e in edges if e["source"] == node_id]
+    if len(outgoing) >= 2:
+        conditions = {e.get("condition", "on_success") for e in outgoing}
+        if len(conditions) > 1 or conditions - {"on_success"}:
+            return "decision"
+
+    # Sub-agent / subprocess nodes → subroutine (double-bordered rect)
+    if node.get("sub_agents"):
+        return "subprocess"
+
+    # Database / data store nodes → cylinder
+    db_tool_hints = {
+        "query_database",
+        "sql_query",
+        "read_table",
+        "write_table",
+        "save_data",
+        "load_data",
+    }
+    db_desc_hints = {"database", "data store", "storage", "persist", "cache"}
+    if node_tools & db_tool_hints or any(h in desc for h in db_desc_hints):
+        return "database"
+
+    # Document generation nodes → document shape
+    doc_tool_hints = {
+        "generate_report",
+        "create_document",
+        "write_report",
+        "render_template",
+        "export_pdf",
+    }
+    doc_desc_hints = {"report", "document", "summary", "write up", "writeup"}
+    if node_tools & doc_tool_hints or any(h in desc for h in doc_desc_hints):
+        return "document"
+
+    # I/O nodes: external data ingestion or delivery → parallelogram
+    io_tool_hints = {
+        "serve_file_to_user",
+        "send_email",
+        "post_message",
+        "upload_file",
+        "download_file",
+        "fetch_url",
+        "post_to_slack",
+        "send_notification",
+        "display_results",
+    }
+    io_desc_hints = {"deliver", "send", "output", "notify", "publish"}
+    if node_tools & io_tool_hints or any(h in desc for h in io_desc_hints):
+        return "io"
+
+    # Default: process (rectangle)
+    return "process"
+
+
+# ── Draft synthesis from runtime graph ───────────────────────────────────────
+
+
+def synthesize_draft_from_runtime(
+    runtime_nodes: list,
+    runtime_edges: list,
+    agent_name: str = "",
+    goal_name: str = "",
+) -> tuple[dict, dict[str, list[str]]]:
+    """Generate a flowchart draft from a loaded runtime graph.
+
+    Used for agents that were never planned through the draft workflow
+    (e.g., hand-coded or loaded from "my agents"). Produces a valid
+    DraftGraph structure with auto-classified flowchart types.
+    """
+    nodes: list[dict] = []
+    edges: list[dict] = []
+    node_ids = {n.id for n in runtime_nodes}
+
+    # Build edge dicts first (needed for classification)
+    for i, re in enumerate(runtime_edges):
+        edges.append(
+            {
+                "id": f"edge-{i}",
+                "source": re.source,
+                "target": re.target,
+                "condition": str(re.condition.value)
+                if hasattr(re.condition, "value")
+                else str(re.condition),
+                "description": getattr(re, "description", "") or "",
+                "label": "",
+            }
+        )
+
+    # Terminal detection — exclude sub-agent nodes (they are leaf helpers, not endpoints)
+    sub_agent_ids: set[str] = set()
+    for rn in runtime_nodes:
+        for sa_id in getattr(rn, "sub_agents", None) or []:
+            sub_agent_ids.add(sa_id)
+    sources = {e["source"] for e in edges}
+    terminal_ids = node_ids - sources - sub_agent_ids
+    if not terminal_ids and runtime_nodes:
+        terminal_ids = {runtime_nodes[-1].id}
+
+    # Build node dicts with classification
+    total = len(runtime_nodes)
+    for i, rn in enumerate(runtime_nodes):
+        node: dict = {
+            "id": rn.id,
+            "name": rn.name,
+            "description": rn.description or "",
+            "node_type": getattr(rn, "node_type", "event_loop") or "event_loop",
+            "tools": list(rn.tools) if rn.tools else [],
+            "input_keys": list(rn.input_keys) if rn.input_keys else [],
+            "output_keys": list(rn.output_keys) if rn.output_keys else [],
+            "success_criteria": getattr(rn, "success_criteria", "") or "",
+            "sub_agents": list(rn.sub_agents) if getattr(rn, "sub_agents", None) else [],
+        }
+        fc_type = classify_flowchart_node(node, i, total, edges, terminal_ids)
+        fc_meta = FLOWCHART_TYPES[fc_type]
+        node["flowchart_type"] = fc_type
+        node["flowchart_shape"] = fc_meta["shape"]
+        node["flowchart_color"] = fc_meta["color"]
+        nodes.append(node)
+
+    # Add visual edges from parent nodes to their sub_agents.
+    # Sub-agents are connected via the sub_agents field, not via EdgeSpec,
+    # so they'd appear as disconnected islands without this.
+    # Two edges per sub-agent: delegate (parent→sub) and report (sub→parent).
+    edge_counter = len(edges)
+    for node in nodes:
+        for sa_id in node.get("sub_agents") or []:
+            if sa_id in node_ids:
+                edges.append(
+                    {
+                        "id": f"edge-subagent-{edge_counter}",
+                        "source": node["id"],
+                        "target": sa_id,
+                        "condition": "always",
+                        "description": "sub-agent delegation",
+                        "label": "delegate",
+                    }
+                )
+                edge_counter += 1
+                edges.append(
+                    {
+                        "id": f"edge-subagent-{edge_counter}",
+                        "source": sa_id,
+                        "target": node["id"],
+                        "condition": "always",
+                        "description": "sub-agent report back",
+                        "label": "report",
+                    }
+                )
+                edge_counter += 1
+
+    # Group sub-agent nodes under their parent in the flowchart map
+    # (mirrors what _dissolve_planning_nodes does for planned drafts)
+    sub_agent_ids_final: set[str] = set()
+    for node in nodes:
+        for sa_id in node.get("sub_agents") or []:
+            if sa_id in node_ids:
+                sub_agent_ids_final.add(sa_id)
+
+    fmap: dict[str, list[str]] = {}
+    for node in nodes:
+        nid = node["id"]
+        if nid in sub_agent_ids_final:
+            continue  # skip — will be included via parent
+        absorbed = [nid]
+        for sa_id in node.get("sub_agents") or []:
+            if sa_id in node_ids:
+                absorbed.append(sa_id)
+        fmap[nid] = absorbed
+
+    draft = {
+        "agent_name": agent_name,
+        "goal": goal_name,
+        "description": "",
+        "success_criteria": [],
+        "constraints": [],
+        "nodes": nodes,
+        "edges": edges,
+        "entry_node": nodes[0]["id"] if nodes else "",
+        "terminal_nodes": sorted(terminal_ids),
+        "flowchart_legend": {
+            fc_type: {"shape": meta["shape"], "color": meta["color"]}
+            for fc_type, meta in FLOWCHART_TYPES.items()
+        },
+    }
+
+    return draft, fmap
+
+
+# ── Fallback generation entry point ──────────────────────────────────────────
+
+
+def generate_fallback_flowchart(
+    graph: Any,
+    goal: Any,
+    agent_path: Path,
+) -> None:
+    """Generate flowchart.json from a runtime GraphSpec if none exists.
+
+    This is a no-op if flowchart.json already exists. On failure, logs a
+    warning but never raises — agent loading must not be blocked by
+    flowchart generation.
+    """
+    try:
+        existing_draft, _ = load_flowchart_file(agent_path)
+        if existing_draft is not None:
+            return  # already have one
+
+        draft, fmap = synthesize_draft_from_runtime(
+            runtime_nodes=list(graph.nodes),
+            runtime_edges=list(graph.edges),
+            agent_name=agent_path.name,
+            goal_name=goal.name if goal else "",
+        )
+
+        # Enrich with Goal metadata
+        if goal:
+            draft["goal"] = goal.description or goal.name or ""
+            draft["success_criteria"] = [sc.description for sc in (goal.success_criteria or [])]
+            draft["constraints"] = [c.description for c in (goal.constraints or [])]
+
+        # Use entry_node/terminal_nodes from GraphSpec if available
+        if graph.entry_node:
+            draft["entry_node"] = graph.entry_node
+        if graph.terminal_nodes:
+            draft["terminal_nodes"] = list(graph.terminal_nodes)
+
+        save_flowchart_file(agent_path, draft, fmap)
+        logger.info("Generated fallback flowchart.json for %s", agent_path.name)
+    except Exception:
+        logger.warning(
+            "Failed to generate fallback flowchart for %s",
+            agent_path,
+            exc_info=True,
+        )

core/framework/tools/queen_memory_tools.py

@@ -1,8 +1,9 @@
-"""Tool for the queen to write to her episodic memory.
+"""Tools for the queen to read and write episodic memory.
 
 The queen can consciously record significant moments during a session — like
-writing in a diary. Semantic memory (MEMORY.md) is updated automatically at
-session end and is never written by the queen directly.
+writing in a diary — and recall past diary entries when needed. Semantic
+memory (MEMORY.md) is updated automatically at session end and is never
+written by the queen directly.
 """
 
 from __future__ import annotations
@@ -33,6 +34,67 @@ def write_to_diary(entry: str) -> str:
     return "Diary entry recorded."
 
 
+def recall_diary(query: str = "", days_back: int = 7) -> str:
+    """Search recent diary entries (episodic memory).
+
+    Use this when the user asks about what happened in the past — "what did we
+    do yesterday?", "what happened last week?", "remind me about the pipeline
+    issue", etc. Also use it proactively when you need context from recent
+    sessions to answer a question or make a decision.
+
+    Args:
+        query: Optional keyword or phrase to filter entries. If empty, all
+            recent entries are returned.
+        days_back: How many days to look back (1–30). Defaults to 7.
+    """
+    from datetime import date, timedelta
+
+    from framework.agents.queen.queen_memory import read_episodic_memory
+
+    days_back = max(1, min(days_back, 30))
+    today = date.today()
+    results: list[str] = []
+    total_chars = 0
+    char_budget = 12_000
+
+    for offset in range(days_back):
+        d = today - timedelta(days=offset)
+        content = read_episodic_memory(d)
+        if not content:
+            continue
+        # If a query is given, only include entries that mention it
+        if query:
+            # Check each section (split by ###) for relevance
+            sections = content.split("### ")
+            matched = [s for s in sections if query.lower() in s.lower()]
+            if not matched:
+                continue
+            content = "### ".join(matched)
+        label = d.strftime("%B %-d, %Y")
+        if d == today:
+            label = f"Today — {label}"
+        entry = f"## {label}\n\n{content}"
+        if total_chars + len(entry) > char_budget:
+            remaining = char_budget - total_chars
+            if remaining > 200:
+                # Fit a partial entry within budget
+                trimmed = content[: remaining - 100] + "\n\n…(truncated)"
+                results.append(f"## {label}\n\n{trimmed}")
+            else:
+                results.append(f"## {label}\n\n(truncated — hit size limit)")
+            break
+        results.append(entry)
+        total_chars += len(entry)
+
+    if not results:
+        if query:
+            return f"No diary entries matching '{query}' in the last {days_back} days."
+        return f"No diary entries found in the last {days_back} days."
+
+    return "\n\n---\n\n".join(results)
+
+
 def register_queen_memory_tools(registry: ToolRegistry) -> None:
-    """Register the episodic memory tool into the queen's tool registry."""
+    """Register the episodic memory tools into the queen's tool registry."""
     registry.register_function(write_to_diary)
+    registry.register_function(recall_diary)

core/framework/tools/session_graph_tools.py

@@ -78,19 +78,6 @@ def register_graph_tools(registry: ToolRegistry, runtime: AgentRuntime) -> int:
                 isolation_level="shared",
             )
 
-        # Async entry points
-        for aep in runner.graph.async_entry_points:
-            entry_points[aep.id] = EntryPointSpec(
-                id=aep.id,
-                name=aep.name,
-                entry_node=aep.entry_node,
-                trigger_type=aep.trigger_type,
-                trigger_config=aep.trigger_config,
-                isolation_level=aep.isolation_level,
-                priority=aep.priority,
-                max_concurrent=aep.max_concurrent,
-            )
-
         await runtime.add_graph(
             graph_id=graph_id,
             graph=runner.graph,

core/framework/tools/worker_monitoring_tools.py

@@ -1,20 +1,17 @@
-"""Worker monitoring tools for the Health Judge and Queen triage agents.
+"""Worker monitoring tools for Queen triage agents.
 
 Three tools are registered by ``register_worker_monitoring_tools()``:
 
 - ``get_worker_health_summary`` — reads the worker's session log files and
   returns a compact health snapshot (recent verdicts, step count, timing).
   session_id is optional: if omitted, the most recent active session is
-  auto-discovered from storage. No agent-side configuration required.
-  Used by the Health Judge on every timer tick.
+  auto-discovered from storage.
 
 - ``emit_escalation_ticket`` — validates and publishes an EscalationTicket
   to the shared EventBus as a WORKER_ESCALATION_TICKET event.
-  Used by the Health Judge when it decides to escalate.
 
 - ``notify_operator`` — emits a QUEEN_INTERVENTION_REQUESTED event so the TUI
   can surface a non-disruptive operator notification.
-  Used by the Queen's ticket_triage_node when it decides to intervene.
 
 Usage::
 
@@ -45,8 +42,9 @@ def register_worker_monitoring_tools(
     registry: ToolRegistry,
     event_bus: EventBus,
     storage_path: Path,
-    stream_id: str = "judge",
+    stream_id: str = "monitoring",
     worker_graph_id: str | None = None,
+    default_session_id: str | None = None,
 ) -> int:
     """Register worker monitoring tools bound to *event_bus* and *storage_path*.
 
@@ -55,9 +53,15 @@ def register_worker_monitoring_tools(
         event_bus: The shared EventBus for the worker runtime.
         storage_path: Root storage path of the worker runtime
                       (e.g. ``~/.hive/agents/{name}``).
-        stream_id: Stream ID used when emitting events; defaults to judge's stream.
+        stream_id: Stream ID used when emitting events.
         worker_graph_id: The primary worker graph's ID. Included in health summary
                          so the judge can populate ticket identity fields accurately.
+        default_session_id: When set, ``get_worker_health_summary`` uses this
+                            session ID as the default instead of auto-discovering
+                            the most-recent-by-mtime session. Callers should pass
+                            the queen's own session ID so that after a cold-restore
+                            the monitoring tool reads the correct worker session
+                            rather than a stale orphaned one.
 
     Returns:
         Number of tools registered.
@@ -65,7 +69,7 @@ def register_worker_monitoring_tools(
     from framework.llm.provider import Tool
 
     storage_path = Path(storage_path)
-    # Derive agent identity from storage path so the judge can fill ticket fields.
+    # Derive agent identity from storage path for ticket fields.
     # storage_path is ~/.hive/agents/{agent_name} — the name is the last component.
     _worker_agent_id: str = storage_path.name
     _worker_graph_id: str = worker_graph_id or storage_path.name
@@ -100,23 +104,29 @@ def register_worker_monitoring_tools(
             if not sessions_dir.exists():
                 return json.dumps({"error": "No sessions found — worker has not started yet"})
 
-            candidates = [
-                d for d in sessions_dir.iterdir() if d.is_dir() and (d / "state.json").exists()
-            ]
-            if not candidates:
-                return json.dumps({"error": "No sessions found — worker has not started yet"})
+            # Prefer the queen's own session ID (set at registration time) over
+            # mtime-based discovery, which can pick a stale orphaned session after
+            # a cold-restore when a newer-but-empty session directory exists.
+            if default_session_id and (sessions_dir / default_session_id).is_dir():
+                session_id = default_session_id
+            else:
+                candidates = [
+                    d for d in sessions_dir.iterdir() if d.is_dir() and (d / "state.json").exists()
+                ]
+                if not candidates:
+                    return json.dumps({"error": "No sessions found — worker has not started yet"})
 
-            def _sort_key(d: Path):
-                try:
-                    state = json.loads((d / "state.json").read_text(encoding="utf-8"))
-                    # in_progress/running sorts before completed/failed
-                    priority = 0 if state.get("status", "") in ("in_progress", "running") else 1
-                    return (priority, -d.stat().st_mtime)
-                except Exception:
-                    return (2, 0)
+                def _sort_key(d: Path):
+                    try:
+                        state = json.loads((d / "state.json").read_text(encoding="utf-8"))
+                        # in_progress/running sorts before completed/failed
+                        priority = 0 if state.get("status", "") in ("in_progress", "running") else 1
+                        return (priority, -d.stat().st_mtime)
+                    except Exception:
+                        return (2, 0)
 
-            candidates.sort(key=_sort_key)
-            session_id = candidates[0].name
+                candidates.sort(key=_sort_key)
+                session_id = candidates[0].name
 
         # Resolve log paths
         session_dir = storage_path / "sessions" / session_id
@@ -201,10 +211,9 @@ def register_worker_monitoring_tools(
         description=(
             "Read the worker agent's execution logs and return a compact health snapshot. "
             "Returns worker_agent_id and worker_graph_id (use these for ticket identity fields), "
-            "recent judge verdicts, step count, time since last step, and "
+            "recent verdicts, step count, time since last step, and "
             "a snippet of the most recent LLM output. "
-            "session_id is optional — omit it to auto-discover the most recent active session. "
-            "Use this on every health check to observe trends."
+            "session_id is optional — omit it to auto-discover the most recent active session."
         ),
         parameters={
             "type": "object",
@@ -241,8 +250,7 @@ def register_worker_monitoring_tools(
         """Validate and publish an EscalationTicket to the shared EventBus.
 
         ticket_json must be a JSON string containing all required EscalationTicket
-        fields. The ticket is validated before publishing — this ensures the judge
-        has genuinely filled out all required evidence fields.
+        fields. The ticket is validated before publishing.
 
         Returns a confirmation JSON with the ticket_id on success, or an error.
         """
@@ -257,7 +265,7 @@ def register_worker_monitoring_tools(
         try:
             await event_bus.emit_worker_escalation_ticket(
                 stream_id=stream_id,
-                node_id="judge",
+                node_id="monitoring",
                 ticket=ticket.model_dump(),
             )
             logger.info(
@@ -280,7 +288,6 @@ def register_worker_monitoring_tools(
         name="emit_escalation_ticket",
         description=(
             "Validate and publish a structured EscalationTicket to the shared EventBus. "
-            "The Queen's ticket_receiver entry point will fire and triage the ticket. "
             "ticket_json must be a JSON string with all required EscalationTicket fields: "
             "worker_agent_id, worker_session_id, worker_node_id, worker_graph_id, "
             "severity (low/medium/high/critical), cause, judge_reasoning, suggested_action, "

core/frontend/src/api/client.ts

@@ -38,4 +38,9 @@ export const api = {
       body: body ? JSON.stringify(body) : undefined,
     }),
   delete: <T>(path: string) => request<T>(path, { method: "DELETE" }),
+  patch: <T>(path: string, body?: unknown) =>
+    request<T>(path, {
+      method: "PATCH",
+      body: body ? JSON.stringify(body) : undefined,
+    }),
 };

core/frontend/src/api/graphs.ts

@@ -1,5 +1,5 @@
 import { api } from "./client";
-import type { GraphTopology, NodeDetail, NodeCriteria, ToolInfo } from "./types";
+import type { GraphTopology, NodeDetail, NodeCriteria, ToolInfo, DraftGraph, FlowchartMap } from "./types";
 
 export const graphsApi = {
   nodes: (sessionId: string, graphId: string, workerSessionId?: string) =>
@@ -26,4 +26,14 @@ export const graphsApi = {
     api.get<{ tools: ToolInfo[] }>(
       `/sessions/${sessionId}/graphs/${graphId}/nodes/${nodeId}/tools`,
     ),
+
+  draftGraph: (sessionId: string) =>
+    api.get<{ draft: DraftGraph | null }>(
+      `/sessions/${sessionId}/draft-graph`,
+    ),
+
+  flowchartMap: (sessionId: string) =>
+    api.get<FlowchartMap>(
+      `/sessions/${sessionId}/flowchart-map`,
+    ),
 };

core/frontend/src/api/sessions.ts

@@ -1,11 +1,11 @@
 import { api } from "./client";
 import type {
+  AgentEvent,
   LiveSession,
   LiveSessionDetail,
   SessionSummary,
   SessionDetail,
   Checkpoint,
-  Message,
   EntryPoint,
 } from "./types";
 
@@ -64,12 +64,22 @@ export const sessionsApi = {
       `/sessions/${sessionId}/entry-points`,
     ),
 
+  updateTrigger: (
+    sessionId: string,
+    triggerId: string,
+    patch: { task?: string; trigger_config?: Record<string, unknown> },
+  ) =>
+    api.patch<{ trigger_id: string; task: string; trigger_config: Record<string, unknown> }>(
+      `/sessions/${sessionId}/triggers/${triggerId}`,
+      patch,
+    ),
+
   graphs: (sessionId: string) =>
     api.get<{ graphs: string[] }>(`/sessions/${sessionId}/graphs`),
 
-  /** Get queen conversation history for a session (works for cold/post-restart sessions too). */
-  queenMessages: (sessionId: string) =>
-    api.get<{ messages: Message[]; session_id: string }>(`/sessions/${sessionId}/queen-messages`),
+  /** Get persisted eventbus log for a session (works for cold sessions — used for full UI replay). */
+  eventsHistory: (sessionId: string) =>
+    api.get<{ events: AgentEvent[]; session_id: string }>(`/sessions/${sessionId}/events/history`),
 
   /** List all queen sessions on disk — live + cold (post-restart). */
   history: () =>
@@ -105,12 +115,4 @@ export const sessionsApi = {
     api.post<{ execution_id: string }>(
       `/sessions/${sessionId}/worker-sessions/${wsId}/checkpoints/${checkpointId}/restore`,
     ),
-
-  messages: (sessionId: string, wsId: string, nodeId?: string) => {
-    const params = new URLSearchParams({ client_only: "true" });
-    if (nodeId) params.set("node_id", nodeId);
-    return api.get<{ messages: Message[] }>(
-      `/sessions/${sessionId}/worker-sessions/${wsId}/messages?${params}`,
-    );
-  },
 };

core/frontend/src/api/types.ts

@@ -31,6 +31,8 @@ export interface EntryPoint {
   entry_node: string;
   trigger_type: string;
   trigger_config?: Record<string, unknown>;
+  /** Worker task string when this trigger fires autonomously. */
+  task?: string;
   /** Seconds until the next timer fire (only present for timer entry points). */
   next_fire_in?: number;
 }
@@ -41,6 +43,7 @@ export interface DiscoverEntry {
   description: string;
   category: string;
   session_count: number;
+  run_count: number;
   node_count: number;
   tool_count: number;
   tags: string[];
@@ -191,6 +194,56 @@ export interface GraphTopology {
   entry_points?: EntryPoint[];
 }
 
+// --- Draft graph types (planning phase) ---
+
+export interface DraftNode {
+  id: string;
+  name: string;
+  description: string;
+  node_type: string;
+  tools: string[];
+  input_keys: string[];
+  output_keys: string[];
+  success_criteria: string;
+  sub_agents: string[];
+  /** For decision nodes: the yes/no question evaluated during dissolution. */
+  decision_clause?: string;
+  flowchart_type: string;
+  flowchart_shape: string;
+  flowchart_color: string;
+}
+
+export interface DraftEdge {
+  id: string;
+  source: string;
+  target: string;
+  condition: string;
+  description: string;
+  /** Short label shown on the flowchart edge (e.g. "Yes", "No"). */
+  label?: string;
+}
+
+export interface DraftGraph {
+  agent_name: string;
+  goal: string;
+  description: string;
+  success_criteria: string[];
+  constraints: string[];
+  nodes: DraftNode[];
+  edges: DraftEdge[];
+  entry_node: string;
+  terminal_nodes: string[];
+  flowchart_legend: Record<string, { shape: string; color: string }>;
+}
+
+/** Mapping from runtime graph nodes → original flowchart draft nodes. */
+export interface FlowchartMap {
+  /** runtime_node_id → list of original draft node IDs it absorbed. */
+  map: Record<string, string[]> | null;
+  /** Original draft graph preserved before planning-node dissolution (decision + subagent). */
+  original_draft: DraftGraph | null;
+}
+
 export interface NodeCriteria {
   node_id: string;
   success_criteria: string | null;
@@ -261,6 +314,7 @@ export type EventTypeName =
   | "tool_call_completed"
   | "client_output_delta"
   | "client_input_requested"
+  | "client_input_received"
   | "node_internal_output"
   | "node_input_blocked"
   | "node_stalled"
@@ -276,7 +330,15 @@ export type EventTypeName =
   | "worker_loaded"
   | "credentials_required"
   | "queen_phase_changed"
-  | "subagent_report";
+  | "subagent_report"
+  | "draft_graph_updated"
+  | "flowchart_map_updated"
+  | "trigger_available"
+  | "trigger_activated"
+  | "trigger_deactivated"
+  | "trigger_fired"
+  | "trigger_removed"
+  | "trigger_updated";
 
 export interface AgentEvent {
   type: EventTypeName;
@@ -287,4 +349,5 @@ export interface AgentEvent {
   timestamp: string;
   correlation_id: string | null;
   graph_id: string | null;
+  run_id?: string | null;
 }

core/frontend/src/components/AgentGraph.tsx

@@ -1,626 +0,0 @@
-import { memo, useMemo, useState, useRef } from "react";
-import { Play, Pause, Loader2, CheckCircle2 } from "lucide-react";
-
-export type NodeStatus = "running" | "complete" | "pending" | "error" | "looping";
-
-export type NodeType = "execution" | "trigger";
-
-export interface GraphNode {
-  id: string;
-  label: string;
-  status: NodeStatus;
-  nodeType?: NodeType;
-  triggerType?: string;
-  triggerConfig?: Record<string, unknown>;
-  next?: string[];
-  backEdges?: string[];
-  iterations?: number;
-  maxIterations?: number;
-  statusLabel?: string;
-  edgeLabels?: Record<string, string>;
-}
-
-type RunState = "idle" | "deploying" | "running";
-
-interface AgentGraphProps {
-  nodes: GraphNode[];
-  title: string;
-  onNodeClick?: (node: GraphNode) => void;
-  onRun?: () => void;
-  onPause?: () => void;
-  version?: string;
-  runState?: RunState;
-  building?: boolean;
-  queenPhase?: "planning" | "building" | "staging" | "running";
-}
-
-// --- Extracted RunButton so hover state survives parent re-renders ---
-interface RunButtonProps {
-  runState: RunState;
-  disabled: boolean;
-  onRun: () => void;
-  onPause: () => void;
-  btnRef: React.Ref<HTMLButtonElement>;
-}
-
-const RunButton = memo(function RunButton({ runState, disabled, onRun, onPause, btnRef }: RunButtonProps) {
-  const [hovered, setHovered] = useState(false);
-  const showPause = runState === "running" && hovered;
-
-  return (
-    <button
-      ref={btnRef}
-      onClick={runState === "running" ? onPause : onRun}
-      disabled={runState === "deploying" || disabled}
-      onMouseEnter={() => setHovered(true)}
-      onMouseLeave={() => setHovered(false)}
-      className={`flex items-center gap-1.5 px-2.5 py-1 rounded-md text-[11px] font-semibold transition-all duration-200 ${
-        showPause
-          ? "bg-amber-500/15 text-amber-400 border border-amber-500/40 hover:bg-amber-500/25 active:scale-95 cursor-pointer"
-          : runState === "running"
-          ? "bg-green-500/15 text-green-400 border border-green-500/30 cursor-pointer"
-          : runState === "deploying"
-          ? "bg-primary/10 text-primary border border-primary/20 cursor-default"
-          : disabled
-          ? "bg-muted/30 text-muted-foreground/40 border border-border/20 cursor-not-allowed"
-          : "bg-primary/10 text-primary border border-primary/20 hover:bg-primary/20 hover:border-primary/40 active:scale-95"
-      }`}
-    >
-      {runState === "deploying" ? (
-        <Loader2 className="w-3 h-3 animate-spin" />
-      ) : showPause ? (
-        <Pause className="w-3 h-3 fill-current" />
-      ) : runState === "running" ? (
-        <CheckCircle2 className="w-3 h-3" />
-      ) : (
-        <Play className="w-3 h-3 fill-current" />
-      )}
-      {runState === "deploying" ? "Deploying\u2026" : showPause ? "Pause" : runState === "running" ? "Running" : "Run"}
-    </button>
-  );
-});
-
-const NODE_W_MAX = 180;
-const NODE_H = 44;
-const GAP_Y = 48;
-const TOP_Y = 30;
-const MARGIN_LEFT = 20;
-const MARGIN_RIGHT = 50; // space for back-edge arcs
-const SVG_BASE_W = 320;
-const GAP_X = 12;
-
-// Unified amber/gold palette
-const statusColors: Record<NodeStatus, { dot: string; bg: string; border: string; glow: string }> = {
-  running: {
-    dot: "hsl(45,95%,58%)",
-    bg: "hsl(45,95%,58%,0.08)",
-    border: "hsl(45,95%,58%,0.5)",
-    glow: "hsl(45,95%,58%,0.15)",
-  },
-  looping: {
-    dot: "hsl(38,90%,55%)",
-    bg: "hsl(38,90%,55%,0.08)",
-    border: "hsl(38,90%,55%,0.5)",
-    glow: "hsl(38,90%,55%,0.15)",
-  },
-  complete: {
-    dot: "hsl(43,70%,45%)",
-    bg: "hsl(43,70%,45%,0.05)",
-    border: "hsl(43,70%,45%,0.25)",
-    glow: "none",
-  },
-  pending: {
-    dot: "hsl(35,15%,28%)",
-    bg: "hsl(35,10%,12%)",
-    border: "hsl(35,10%,20%)",
-    glow: "none",
-  },
-  error: {
-    dot: "hsl(0,65%,55%)",
-    bg: "hsl(0,65%,55%,0.06)",
-    border: "hsl(0,65%,55%,0.3)",
-    glow: "hsl(0,65%,55%,0.1)",
-  },
-};
-
-// Trigger node palette — cool blue-gray, visually distinct from amber execution nodes
-const triggerColors = {
-  bg: "hsl(210,25%,14%)",
-  border: "hsl(210,30%,30%)",
-  text: "hsl(210,30%,65%)",
-  icon: "hsl(210,40%,55%)",
-};
-
-const triggerIcons: Record<string, string> = {
-  webhook: "\u26A1",  // lightning bolt
-  timer: "\u23F1",    // stopwatch
-  api: "\u2192",      // right arrow
-  event: "\u223F",    // sine wave
-};
-
-/** Truncate label to fit within `availablePx` at the given fontSize. */
-function truncateLabel(label: string, availablePx: number, fontSize: number): string {
-  const avgCharW = fontSize * 0.58;
-  const maxChars = Math.floor(availablePx / avgCharW);
-  if (label.length <= maxChars) return label;
-  return label.slice(0, Math.max(maxChars - 1, 1)) + "\u2026";
-}
-
-export default function AgentGraph({ nodes, title: _title, onNodeClick, onRun, onPause, version, runState: externalRunState, building, queenPhase }: AgentGraphProps) {
-  const [localRunState, setLocalRunState] = useState<RunState>("idle");
-  const runState = externalRunState ?? localRunState;
-  const runBtnRef = useRef<HTMLButtonElement>(null);
-
-  const handleRun = () => {
-    if (runState !== "idle") return;
-    if (onRun) {
-      onRun();
-    } else {
-      setLocalRunState("deploying");
-      setTimeout(() => setLocalRunState("running"), 1800);
-      setTimeout(() => setLocalRunState("idle"), 5000);
-    }
-  };
-
-  const idxMap = useMemo(() => Object.fromEntries(nodes.map((n, i) => [n.id, i])), [nodes]);
-
-  const backEdges = useMemo(() => {
-    const edges: { fromIdx: number; toIdx: number }[] = [];
-    nodes.forEach((n, i) => {
-      (n.next || []).forEach((toId) => {
-        const toIdx = idxMap[toId];
-        if (toIdx !== undefined && toIdx <= i) edges.push({ fromIdx: i, toIdx });
-      });
-      (n.backEdges || []).forEach((toId) => {
-        const toIdx = idxMap[toId];
-        if (toIdx !== undefined) edges.push({ fromIdx: i, toIdx });
-      });
-    });
-    return edges;
-  }, [nodes, idxMap]);
-
-  const forwardEdges = useMemo(() => {
-    const edges: { fromIdx: number; toIdx: number; fanCount: number; fanIndex: number; label?: string }[] = [];
-    nodes.forEach((n, i) => {
-      const targets = (n.next || [])
-        .map((toId) => ({ toId, toIdx: idxMap[toId] }))
-        .filter((t): t is { toId: string; toIdx: number } => t.toIdx !== undefined && t.toIdx > i);
-      targets.forEach(({ toId, toIdx }, fi) => {
-        edges.push({
-          fromIdx: i,
-          toIdx,
-          fanCount: targets.length,
-          fanIndex: fi,
-          label: n.edgeLabels?.[toId],
-        });
-      });
-    });
-    return edges;
-  }, [nodes, idxMap]);
-
-  // --- Layer-based layout computation ---
-  const layout = useMemo(() => {
-    if (nodes.length === 0) {
-      return { layers: [] as number[], cols: [] as number[], maxCols: 1, nodeW: NODE_W_MAX, colSpacing: 0, firstColX: MARGIN_LEFT };
-    }
-
-    // 1. Build reverse adjacency from forward edges (who are the parents of each node)
-    const parents = new Map<number, number[]>();
-    nodes.forEach((_, i) => parents.set(i, []));
-    forwardEdges.forEach((e) => {
-      parents.get(e.toIdx)!.push(e.fromIdx);
-    });
-
-    // 2. Assign layers via longest-path from entry
-    const layers = new Array(nodes.length).fill(0);
-    for (let i = 0; i < nodes.length; i++) {
-      const pars = parents.get(i) || [];
-      if (pars.length > 0) {
-        layers[i] = Math.max(...pars.map((p) => layers[p])) + 1;
-      }
-    }
-
-    // 3. Group nodes by layer
-    const layerGroups = new Map<number, number[]>();
-    layers.forEach((l, i) => {
-      const group = layerGroups.get(l) || [];
-      group.push(i);
-      layerGroups.set(l, group);
-    });
-
-    // 4. Compute max columns and dynamic node width
-    let maxCols = 1;
-    layerGroups.forEach((group) => {
-      maxCols = Math.max(maxCols, group.length);
-    });
-
-    const usableW = SVG_BASE_W - MARGIN_LEFT - MARGIN_RIGHT;
-    const nodeW = Math.min(NODE_W_MAX, Math.floor((usableW - (maxCols - 1) * GAP_X) / maxCols));
-    const colSpacing = nodeW + GAP_X;
-    const totalNodesW = maxCols * nodeW + (maxCols - 1) * GAP_X;
-    const firstColX = MARGIN_LEFT + (usableW - totalNodesW) / 2;
-
-    // 5. Assign columns within each layer (centered, ordered by parent column)
-    const cols = new Array(nodes.length).fill(0);
-    layerGroups.forEach((group) => {
-      if (group.length === 1) {
-        // Center single node: place at middle column
-        cols[group[0]] = (maxCols - 1) / 2;
-      } else {
-        // Sort group by average parent column to reduce crossings
-        const sorted = [...group].sort((a, b) => {
-          const aParents = parents.get(a) || [];
-          const bParents = parents.get(b) || [];
-          const aAvg = aParents.length > 0 ? aParents.reduce((s, p) => s + cols[p], 0) / aParents.length : 0;
-          const bAvg = bParents.length > 0 ? bParents.reduce((s, p) => s + cols[p], 0) / bParents.length : 0;
-          return aAvg - bAvg;
-        });
-        // Spread evenly, centered within maxCols
-        const offset = (maxCols - group.length) / 2;
-        sorted.forEach((nodeIdx, i) => {
-          cols[nodeIdx] = offset + i;
-        });
-      }
-    });
-
-    return { layers, cols, maxCols, nodeW, colSpacing, firstColX };
-  }, [nodes, forwardEdges]);
-
-  if (nodes.length === 0) {
-    return (
-      <div className="flex flex-col h-full">
-        <div className="px-5 pt-4 pb-2 flex items-center justify-between">
-          <div className="flex items-center gap-2">
-            <p className="text-[11px] text-muted-foreground font-medium uppercase tracking-wider">Pipeline</p>
-            {version && (
-              <span className="text-[10px] font-mono font-medium text-muted-foreground/60 border border-border/30 rounded px-1 py-0.5 leading-none">
-                {version}
-              </span>
-            )}
-          </div>
-          <RunButton runState={runState} disabled={nodes.length === 0 || queenPhase === "building" || queenPhase === "planning"} onRun={handleRun} onPause={onPause ?? (() => {})} btnRef={runBtnRef} />
-        </div>
-        <div className="flex-1 flex items-center justify-center px-5">
-          {building ? (
-            <div className="flex flex-col items-center gap-3">
-              <Loader2 className="w-6 h-6 animate-spin text-primary/60" />
-              <p className="text-xs text-muted-foreground/80 text-center">Building agent...</p>
-            </div>
-          ) : (
-            <p className="text-xs text-muted-foreground/60 text-center italic">No pipeline configured yet.<br/>Chat with the Queen to get started.</p>
-          )}
-        </div>
-      </div>
-    );
-  }
-
-  const { layers, cols, nodeW, colSpacing, firstColX } = layout;
-
-  const nodePos = (i: number) => ({
-    x: firstColX + cols[i] * colSpacing,
-    y: TOP_Y + layers[i] * (NODE_H + GAP_Y),
-  });
-
-  const maxLayer = nodes.length > 0 ? Math.max(...layers) : 0;
-  const svgHeight = TOP_Y * 2 + (maxLayer + 1) * NODE_H + maxLayer * GAP_Y + 10;
-  const backEdgeSpace = backEdges.length > 0 ? MARGIN_RIGHT + backEdges.length * 18 : 20;
-  const svgWidth = Math.max(SVG_BASE_W, firstColX + layout.maxCols * nodeW + (layout.maxCols - 1) * GAP_X + backEdgeSpace);
-
-  // Check if a skip-level forward edge would collide with intermediate nodes
-  const hasCollision = (fromLayer: number, toLayer: number, fromX: number, toX: number): boolean => {
-    const minX = Math.min(fromX, toX);
-    const maxX = Math.max(fromX, toX) + nodeW;
-    for (let i = 0; i < nodes.length; i++) {
-      const l = layers[i];
-      if (l > fromLayer && l < toLayer) {
-        const nx = firstColX + cols[i] * colSpacing;
-        // Check horizontal overlap
-        if (nx < maxX && nx + nodeW > minX) return true;
-      }
-    }
-    return false;
-  };
-
-  const renderForwardEdge = (edge: { fromIdx: number; toIdx: number; fanCount: number; fanIndex: number; label?: string }, i: number) => {
-    const from = nodePos(edge.fromIdx);
-    const to = nodePos(edge.toIdx);
-    const fromCenterX = from.x + nodeW / 2;
-    const toCenterX = to.x + nodeW / 2;
-    const y1 = from.y + NODE_H;
-    const y2 = to.y;
-
-    // Fan-out: spread exit points across the source node's bottom
-    let startX = fromCenterX;
-    if (edge.fanCount > 1) {
-      const spread = nodeW * 0.5;
-      const step = edge.fanCount > 1 ? spread / (edge.fanCount - 1) : 0;
-      startX = fromCenterX - spread / 2 + edge.fanIndex * step;
-    }
-
-    const midY = (y1 + y2) / 2;
-    const fromLayer = layers[edge.fromIdx];
-    const toLayer = layers[edge.toIdx];
-    const skipsLayers = toLayer - fromLayer > 1;
-
-    let d: string;
-    if (skipsLayers && hasCollision(fromLayer, toLayer, from.x, to.x)) {
-      // Route around intermediate nodes: curve to the left
-      const detourX = Math.min(from.x, to.x) - nodeW * 0.4;
-      d = `M ${startX} ${y1} C ${startX} ${y1 + 20}, ${detourX} ${y1 + 20}, ${detourX} ${midY} S ${toCenterX} ${y2 - 20} ${toCenterX} ${y2}`;
-    } else {
-      // Standard bezier: from source bottom to target top
-      d = `M ${startX} ${y1} C ${startX} ${midY}, ${toCenterX} ${midY}, ${toCenterX} ${y2}`;
-    }
-
-    const fromNode = nodes[edge.fromIdx];
-    const isActive = fromNode.status === "complete" || fromNode.status === "running" || fromNode.status === "looping";
-    const strokeColor = isActive ? "hsl(43,70%,45%,0.35)" : "hsl(35,10%,20%)";
-    const arrowColor = isActive ? "hsl(43,70%,45%,0.5)" : "hsl(35,10%,22%)";
-
-    return (
-      <g key={`fwd-${i}`}>
-        <path d={d} fill="none" stroke={strokeColor} strokeWidth={1.5} />
-        <polygon
-          points={`${toCenterX - 4},${y2 - 6} ${toCenterX + 4},${y2 - 6} ${toCenterX},${y2 - 1}`}
-          fill={arrowColor}
-        />
-        {edge.label && (
-          <text
-            x={(startX + toCenterX) / 2 + 8}
-            y={midY - 2}
-            fill="hsl(35,15%,40%)"
-            fontSize={9}
-            fontStyle="italic"
-          >
-            {edge.label}
-          </text>
-        )}
-      </g>
-    );
-  };
-
-  const renderBackEdge = (edge: { fromIdx: number; toIdx: number }, i: number) => {
-    const from = nodePos(edge.fromIdx);
-    const to = nodePos(edge.toIdx);
-
-    const rightX = Math.max(from.x, to.x) + nodeW;
-    const rightOffset = 28 + i * 18;
-    const startX = from.x + nodeW;
-    const startY = from.y + NODE_H / 2;
-    const endX = to.x + nodeW;
-    const endY = to.y + NODE_H / 2;
-    const curveX = rightX + rightOffset;
-    const r = 12;
-
-    const fromNode = nodes[edge.fromIdx];
-    const isActive = fromNode.status === "complete" || fromNode.status === "running" || fromNode.status === "looping";
-    const color = isActive ? "hsl(38,80%,50%,0.3)" : "hsl(35,10%,20%)";
-
-    // Bezier curve with rounded corners
-    const path = `M ${startX} ${startY} C ${startX + r} ${startY}, ${curveX} ${startY}, ${curveX} ${startY - r} L ${curveX} ${endY + r} C ${curveX} ${endY}, ${endX + r} ${endY}, ${endX + 6} ${endY}`;
-
-    return (
-      <g key={`back-${i}`}>
-        <path d={path} fill="none" stroke={color} strokeWidth={1.5} strokeDasharray="4 3" />
-        <polygon
-          points={`${endX + 6},${endY - 3} ${endX + 6},${endY + 3} ${endX},${endY}`}
-          fill={isActive ? "hsl(38,80%,50%,0.45)" : "hsl(35,10%,22%)"}
-        />
-      </g>
-    );
-  };
-
-  const renderTriggerNode = (node: GraphNode, i: number) => {
-    const pos = nodePos(i);
-    const icon = triggerIcons[node.triggerType || ""] || "\u26A1";
-    const triggerFontSize = nodeW < 140 ? 10.5 : 11.5;
-    const triggerAvailW = nodeW - 38;
-    const triggerDisplayLabel = truncateLabel(node.label, triggerAvailW, triggerFontSize);
-    const nextFireIn = node.triggerConfig?.next_fire_in as number | undefined;
-
-    // Format countdown for display below node
-    let countdownLabel: string | null = null;
-    if (nextFireIn != null && nextFireIn > 0) {
-      const h = Math.floor(nextFireIn / 3600);
-      const m = Math.floor((nextFireIn % 3600) / 60);
-      const s = Math.floor(nextFireIn % 60);
-      countdownLabel = h > 0
-        ? `next in ${h}h ${String(m).padStart(2, "0")}m`
-        : `next in ${m}m ${String(s).padStart(2, "0")}s`;
-    }
-
-    return (
-      <g key={node.id} onClick={() => onNodeClick?.(node)} style={{ cursor: onNodeClick ? "pointer" : "default" }}>
-        <title>{node.label}</title>
-        {/* Pill-shaped background with dashed border */}
-        <rect
-          x={pos.x} y={pos.y}
-          width={nodeW} height={NODE_H}
-          rx={NODE_H / 2}
-          fill={triggerColors.bg}
-          stroke={triggerColors.border}
-          strokeWidth={1}
-          strokeDasharray="4 2"
-        />
-
-        {/* Trigger type icon */}
-        <text
-          x={pos.x + 18} y={pos.y + NODE_H / 2}
-          fill={triggerColors.icon} fontSize={13}
-          textAnchor="middle" dominantBaseline="middle"
-        >
-          {icon}
-        </text>
-
-        {/* Label */}
-        <text
-          x={pos.x + 32} y={pos.y + NODE_H / 2}
-          fill={triggerColors.text}
-          fontSize={triggerFontSize}
-          fontWeight={500}
-          dominantBaseline="middle"
-          letterSpacing="0.01em"
-        >
-          {triggerDisplayLabel}
-        </text>
-
-        {/* Countdown label below node */}
-        {countdownLabel && (
-          <text
-            x={pos.x + nodeW / 2} y={pos.y + NODE_H + 13}
-            fill="hsl(210,30%,50%)" fontSize={9.5}
-            textAnchor="middle" fontStyle="italic" opacity={0.7}
-          >
-            {countdownLabel}
-          </text>
-        )}
-      </g>
-    );
-  };
-
-  const renderNode = (node: GraphNode, i: number) => {
-    if (node.nodeType === "trigger") return renderTriggerNode(node, i);
-
-    const pos = nodePos(i);
-    const isActive = node.status === "running" || node.status === "looping";
-    const isDone = node.status === "complete";
-    const colors = statusColors[node.status];
-
-    const fontSize = nodeW < 140 ? 10.5 : 12.5;
-    const labelAvailW = nodeW - 38;
-    const displayLabel = truncateLabel(node.label, labelAvailW, fontSize);
-
-    return (
-      <g key={node.id} onClick={() => onNodeClick?.(node)} style={{ cursor: onNodeClick ? "pointer" : "default" }}>
-        <title>{node.label}</title>
-        {/* Ambient glow for active nodes */}
-        {isActive && (
-          <>
-            <rect
-              x={pos.x - 4} y={pos.y - 4}
-              width={nodeW + 8} height={NODE_H + 8}
-              rx={16} fill={colors.glow}
-            />
-            <rect
-              x={pos.x - 2} y={pos.y - 2}
-              width={nodeW + 4} height={NODE_H + 4}
-              rx={14} fill="none" stroke={colors.dot} strokeWidth={1} opacity={0.25}
-              style={{ animation: "pulse-ring 2.5s ease-out infinite" }}
-            />
-          </>
-        )}
-
-        {/* Node background */}
-        <rect
-          x={pos.x} y={pos.y}
-          width={nodeW} height={NODE_H}
-          rx={12}
-          fill={colors.bg}
-          stroke={colors.border}
-          strokeWidth={isActive ? 1.5 : 1}
-        />
-
-        {/* Status dot */}
-        <circle cx={pos.x + 18} cy={pos.y + NODE_H / 2} r={4.5} fill={colors.dot} />
-        {isActive && (
-          <circle cx={pos.x + 18} cy={pos.y + NODE_H / 2} r={7} fill="none" stroke={colors.dot} strokeWidth={1} opacity={0.3}>
-            <animate attributeName="r" values="7;11;7" dur="2s" repeatCount="indefinite" />
-            <animate attributeName="opacity" values="0.3;0;0.3" dur="2s" repeatCount="indefinite" />
-          </circle>
-        )}
-
-        {/* Check mark for complete */}
-        {isDone && (
-          <text
-            x={pos.x + 18} y={pos.y + NODE_H / 2 + 1}
-            fill={colors.dot} fontSize={8} fontWeight={700}
-            textAnchor="middle" dominantBaseline="middle"
-          >
-            &#x2713;
-          </text>
-        )}
-
-        {/* Label -- truncated with ellipsis for narrow nodes */}
-        <text
-          x={pos.x + 32} y={pos.y + NODE_H / 2}
-          fill={isActive ? "hsl(45,90%,85%)" : isDone ? "hsl(40,20%,75%)" : "hsl(35,10%,45%)"}
-          fontSize={fontSize}
-          fontWeight={isActive ? 600 : isDone ? 500 : 400}
-          dominantBaseline="middle"
-          letterSpacing="0.01em"
-        >
-          {displayLabel}
-        </text>
-
-        {/* Status label for active nodes */}
-        {node.statusLabel && isActive && (
-          <text
-            x={pos.x + nodeW + 10} y={pos.y + NODE_H / 2}
-            fill="hsl(45,80%,60%)" fontSize={10.5} fontStyle="italic"
-            dominantBaseline="middle" opacity={0.8}
-          >
-            {node.statusLabel}
-          </text>
-        )}
-
-        {/* Iteration badge */}
-        {node.iterations !== undefined && node.iterations > 0 && (
-          <g>
-            <rect
-              x={pos.x + nodeW - 36} y={pos.y + NODE_H / 2 - 8}
-              width={26} height={16} rx={8}
-              fill={colors.dot} opacity={0.15}
-            />
-            <text
-              x={pos.x + nodeW - 23} y={pos.y + NODE_H / 2}
-              fill={colors.dot} fontSize={9} fontWeight={600}
-              textAnchor="middle" dominantBaseline="middle" opacity={0.8}
-            >
-              {node.iterations}{node.maxIterations ? `/${node.maxIterations}` : "\u00d7"}
-            </text>
-          </g>
-        )}
-      </g>
-    );
-  };
-
-  return (
-    <div className="flex flex-col h-full">
-      {/* Compact sub-label */}
-      <div className="px-5 pt-4 pb-2 flex items-center justify-between">
-        <div className="flex items-center gap-2">
-          <p className="text-[11px] text-muted-foreground font-medium uppercase tracking-wider">Pipeline</p>
-          {version && (
-            <span className="text-[10px] font-mono font-medium text-muted-foreground/60 border border-border/30 rounded px-1 py-0.5 leading-none">
-              {version}
-            </span>
-          )}
-        </div>
-        <RunButton runState={runState} disabled={nodes.length === 0} onRun={handleRun} onPause={onPause ?? (() => {})} btnRef={runBtnRef} />
-      </div>
-
-      {/* Graph */}
-      <div className="flex-1 overflow-y-auto overflow-x-hidden px-3 pb-5 relative">
-        <svg
-          width={svgWidth}
-          height={svgHeight}
-          viewBox={`0 0 ${svgWidth} ${svgHeight}`}
-          className={`select-none${building ? " opacity-30" : ""}`}
-          style={{ fontFamily: "'Inter', system-ui, sans-serif" }}
-        >
-          {forwardEdges.map((e, i) => renderForwardEdge(e, i))}
-          {backEdges.map((e, i) => renderBackEdge(e, i))}
-          {nodes.map((n, i) => renderNode(n, i))}
-        </svg>
-        {building && (
-          <div className="absolute inset-0 flex items-center justify-center">
-            <div className="flex flex-col items-center gap-3">
-              <Loader2 className="w-6 h-6 animate-spin text-primary/60" />
-              <p className="text-xs text-muted-foreground/80">Rebuilding agent...</p>
-            </div>
-          </div>
-        )}
-      </div>
-    </div>
-  );
-}

core/frontend/src/components/ChatPanel.tsx

@@ -2,6 +2,7 @@ import { memo, useState, useRef, useEffect } from "react";
 import { Send, Square, Crown, Cpu, Check, Loader2 } from "lucide-react";
 import MarkdownContent from "@/components/MarkdownContent";
 import QuestionWidget from "@/components/QuestionWidget";
+import MultiQuestionWidget from "@/components/MultiQuestionWidget";
 
 export interface ChatMessage {
   id: string;
@@ -9,12 +10,14 @@ export interface ChatMessage {
   agentColor: string;
   content: string;
   timestamp: string;
-  type?: "system" | "agent" | "user" | "tool_status" | "worker_input_request";
+  type?: "system" | "agent" | "user" | "tool_status" | "worker_input_request" | "run_divider";
   role?: "queen" | "worker";
   /** Which worker thread this message belongs to (worker agent name) */
   thread?: string;
   /** Epoch ms when this message was first created — used for ordering queen/worker interleaving */
   createdAt?: number;
+  /** Queen phase active when this message was created */
+  phase?: "planning" | "building" | "staging" | "running";
 }
 
 interface ChatPanelProps {
@@ -34,8 +37,12 @@ interface ChatPanelProps {
   pendingQuestion?: string | null;
   /** Options for the pending question */
   pendingOptions?: string[] | null;
+  /** Multiple questions from ask_user_multiple */
+  pendingQuestions?: { id: string; prompt: string; options?: string[] }[] | null;
   /** Called when user submits an answer to the pending question */
   onQuestionSubmit?: (answer: string, isOther: boolean) => void;
+  /** Called when user submits answers to multiple questions */
+  onMultiQuestionSubmit?: (answers: Record<string, string>) => void;
   /** Called when user dismisses the pending question without answering */
   onQuestionDismiss?: () => void;
   /** Queen operating phase — shown as a tag on queen messages */
@@ -149,6 +156,18 @@ const MessageBubble = memo(function MessageBubble({ msg, queenPhase }: { msg: Ch
   const isQueen = msg.role === "queen";
   const color = getColor(msg.agent, msg.role);
 
+  if (msg.type === "run_divider") {
+    return (
+      <div className="flex items-center gap-3 py-2 my-1">
+        <div className="flex-1 h-px bg-border/60" />
+        <span className="text-[10px] text-muted-foreground font-medium uppercase tracking-wider">
+          {msg.content}
+        </span>
+        <div className="flex-1 h-px bg-border/60" />
+      </div>
+    );
+  }
+
   if (msg.type === "system") {
     return (
       <div className="flex justify-center py-1">
@@ -200,13 +219,13 @@ const MessageBubble = memo(function MessageBubble({ msg, queenPhase }: { msg: Ch
             }`}
           >
             {isQueen
-              ? queenPhase === "running"
-                ? "running phase"
-                : queenPhase === "staging"
-                  ? "staging phase"
-                  : queenPhase === "planning"
-                    ? "planning phase"
-                    : "building phase"
+              ? ((msg.phase ?? queenPhase) === "running"
+                ? "running"
+                : (msg.phase ?? queenPhase) === "staging"
+                  ? "staging"
+                  : (msg.phase ?? queenPhase) === "planning"
+                    ? "planning"
+                    : "building")
               : "Worker"}
           </span>
         </div>
@@ -220,9 +239,9 @@ const MessageBubble = memo(function MessageBubble({ msg, queenPhase }: { msg: Ch
       </div>
     </div>
   );
-}, (prev, next) => prev.msg.id === next.msg.id && prev.msg.content === next.msg.content && prev.queenPhase === next.queenPhase);
+}, (prev, next) => prev.msg.id === next.msg.id && prev.msg.content === next.msg.content && prev.msg.phase === next.msg.phase && prev.queenPhase === next.queenPhase);
 
-export default function ChatPanel({ messages, onSend, isWaiting, isWorkerWaiting, isBusy, activeThread, disabled, onCancel, pendingQuestion, pendingOptions, onQuestionSubmit, onQuestionDismiss, queenPhase }: ChatPanelProps) {
+export default function ChatPanel({ messages, onSend, isWaiting, isWorkerWaiting, isBusy, activeThread, disabled, onCancel, pendingQuestion, pendingOptions, pendingQuestions, onQuestionSubmit, onMultiQuestionSubmit, onQuestionDismiss, queenPhase }: ChatPanelProps) {
   const [input, setInput] = useState("");
   const [readMap, setReadMap] = useState<Record<string, number>>({});
   const bottomRef = useRef<HTMLDivElement>(null);
@@ -232,7 +251,13 @@ export default function ChatPanel({ messages, onSend, isWaiting, isWorkerWaiting
 
   const threadMessages = messages.filter((m) => {
     if (m.type === "system" && !m.thread) return false;
-    return m.thread === activeThread;
+    if (m.thread !== activeThread) return false;
+    // Hide queen messages whose content is whitespace-only — these are
+    // tool-use-only turns that have no visible text.  During live operation
+    // tool pills provide context, but on resume the pills are gone so
+    // the empty bubble is meaningless.
+    if (m.role === "queen" && !m.type && (!m.content || !m.content.trim())) return false;
+    return true;
   });
 
   // Mark current thread as read
@@ -332,7 +357,13 @@ export default function ChatPanel({ messages, onSend, isWaiting, isWorkerWaiting
       </div>
 
       {/* Input area — question widget replaces textarea when a question is pending */}
-      {pendingQuestion && pendingOptions && onQuestionSubmit ? (
+      {pendingQuestions && pendingQuestions.length >= 2 && onMultiQuestionSubmit ? (
+        <MultiQuestionWidget
+          questions={pendingQuestions}
+          onSubmit={onMultiQuestionSubmit}
+          onDismiss={onQuestionDismiss}
+        />
+      ) : pendingQuestion && pendingOptions && onQuestionSubmit ? (
         <QuestionWidget
           question={pendingQuestion}
           options={pendingOptions}

core/frontend/src/components/CredentialsModal.tsx

@@ -126,8 +126,13 @@ export default function CredentialsModal({
         // No real path — no credentials to show
         setRows([]);
       }
-    } catch {
-      // Backend unavailable — fall back to legacy props or empty
+    } catch (err) {
+      // Surface the error so the modal shows a meaningful message
+      const message =
+        err instanceof Error ? err.message : "Failed to check credentials";
+      setError(message);
+
+      // Fall back to legacy props or empty rows
       if (legacyCredentials) {
         setRows(legacyCredentials.map(c => ({
           ...c,
@@ -289,11 +294,18 @@ export default function CredentialsModal({
           {/* Status banner */}
           {!loading && (
             <div className={`mx-5 mt-4 px-3 py-2.5 rounded-lg border text-xs font-medium flex items-center gap-2 ${
-              allRequiredMet
-                ? "bg-emerald-500/10 border-emerald-500/20 text-emerald-600"
-                : "bg-destructive/5 border-destructive/20 text-destructive"
+              error && rows.length === 0
+                ? "bg-destructive/5 border-destructive/20 text-destructive"
+                : allRequiredMet
+                  ? "bg-emerald-500/10 border-emerald-500/20 text-emerald-600"
+                  : "bg-destructive/5 border-destructive/20 text-destructive"
             }`}>
-              {allRequiredMet ? (
+              {error && rows.length === 0 ? (
+                <>
+                  <AlertCircle className="w-3.5 h-3.5 flex-shrink-0" />
+                  <span className="break-words">Failed to check credentials: {error}</span>
+                </>
+              ) : allRequiredMet ? (
                 <>
                   <Shield className="w-3.5 h-3.5" />
                   {rows.length === 0

core/frontend/src/components/MultiQuestionWidget.tsx

@@ -0,0 +1,215 @@
+import { useState, useRef, useEffect, useCallback } from "react";
+import { Send, MessageCircleQuestion, X } from "lucide-react";
+
+export interface QuestionItem {
+  id: string;
+  prompt: string;
+  options?: string[];
+}
+
+export interface MultiQuestionWidgetProps {
+  questions: QuestionItem[];
+  onSubmit: (answers: Record<string, string>) => void;
+  onDismiss?: () => void;
+}
+
+export default function MultiQuestionWidget({ questions, onSubmit, onDismiss }: MultiQuestionWidgetProps) {
+  // Per-question state: selected index (null = nothing, options.length = "Other")
+  const [selections, setSelections] = useState<(number | null)[]>(
+    () => questions.map(() => null),
+  );
+  const [customTexts, setCustomTexts] = useState<string[]>(
+    () => questions.map(() => ""),
+  );
+  const [submitted, setSubmitted] = useState(false);
+  const containerRef = useRef<HTMLDivElement>(null);
+
+  // Scroll the first unanswered question into view when it changes
+  useEffect(() => {
+    containerRef.current?.scrollTo({ top: 0, behavior: "smooth" });
+  }, []);
+
+  const canSubmit = questions.every((q, i) => {
+    const sel = selections[i];
+    if (sel === null) return false;
+    const isOther = q.options ? sel === q.options.length : true;
+    if (isOther && !customTexts[i].trim()) return false;
+    return true;
+  });
+
+  const handleSubmit = useCallback(() => {
+    if (!canSubmit || submitted) return;
+    setSubmitted(true);
+    const answers: Record<string, string> = {};
+    for (let i = 0; i < questions.length; i++) {
+      const q = questions[i];
+      const sel = selections[i]!;
+      const isOther = q.options ? sel === q.options.length : true;
+      answers[q.id] = isOther ? customTexts[i].trim() : q.options![sel];
+    }
+    onSubmit(answers);
+  }, [canSubmit, submitted, questions, selections, customTexts, onSubmit]);
+
+  // Enter to submit (only when not focused on a text input)
+  useEffect(() => {
+    const handleKeyDown = (e: KeyboardEvent) => {
+      if (submitted) return;
+      const target = e.target as HTMLElement;
+      const inInput = target.tagName === "INPUT" || target.tagName === "TEXTAREA";
+      if (e.key === "Enter" && !e.shiftKey && !inInput) {
+        e.preventDefault();
+        handleSubmit();
+      }
+    };
+    window.addEventListener("keydown", handleKeyDown);
+    return () => window.removeEventListener("keydown", handleKeyDown);
+  }, [handleSubmit, submitted]);
+
+  if (submitted) return null;
+
+  const answeredCount = selections.filter((s) => s !== null).length;
+
+  return (
+    <div className="p-4">
+      <div className="bg-card border border-border rounded-xl shadow-sm overflow-hidden">
+        {/* Header */}
+        <div className="px-5 pt-4 pb-2 flex items-center gap-3">
+          <div className="w-7 h-7 rounded-lg bg-primary/10 border border-primary/20 flex items-center justify-center flex-shrink-0">
+            <MessageCircleQuestion className="w-3.5 h-3.5 text-primary" />
+          </div>
+          <div className="flex-1 min-w-0">
+            <p className="text-sm font-medium text-foreground">
+              {questions.length} questions
+            </p>
+            <p className="text-[11px] text-muted-foreground">
+              {answeredCount}/{questions.length} answered
+            </p>
+          </div>
+          {onDismiss && (
+            <button
+              onClick={onDismiss}
+              className="p-1 rounded-md text-muted-foreground hover:text-foreground hover:bg-muted/60 transition-colors flex-shrink-0"
+            >
+              <X className="w-4 h-4" />
+            </button>
+          )}
+        </div>
+
+        {/* Questions */}
+        <div
+          ref={containerRef}
+          className="px-5 pb-3 space-y-4 max-h-[400px] overflow-y-auto"
+        >
+          {questions.map((q, qi) => {
+            const sel = selections[qi];
+            const hasOptions = q.options && q.options.length >= 2;
+            const otherIndex = hasOptions ? q.options!.length : 0;
+            const isOtherSelected = sel === otherIndex;
+
+            return (
+              <div key={q.id} className="space-y-1.5">
+                <p className="text-sm font-medium text-foreground">
+                  <span className="text-xs text-muted-foreground mr-1.5">
+                    {qi + 1}.
+                  </span>
+                  {q.prompt}
+                </p>
+
+                {hasOptions ? (
+                  <>
+                    {q.options!.map((opt, oi) => (
+                      <button
+                        key={oi}
+                        onClick={() => {
+                          setSelections((prev) => {
+                            const next = [...prev];
+                            next[qi] = oi;
+                            return next;
+                          });
+                        }}
+                        className={`w-full text-left px-4 py-2 rounded-lg border text-sm transition-colors ${
+                          sel === oi
+                            ? "border-primary bg-primary/10 text-foreground"
+                            : "border-border/60 bg-muted/20 text-foreground hover:border-primary/40 hover:bg-muted/40"
+                        }`}
+                      >
+                        {opt}
+                      </button>
+                    ))}
+                    <input
+                      type="text"
+                      value={customTexts[qi]}
+                      onFocus={() => {
+                        setSelections((prev) => {
+                          const next = [...prev];
+                          next[qi] = otherIndex;
+                          return next;
+                        });
+                      }}
+                      onChange={(e) => {
+                        setSelections((prev) => {
+                          const next = [...prev];
+                          next[qi] = otherIndex;
+                          return next;
+                        });
+                        setCustomTexts((prev) => {
+                          const next = [...prev];
+                          next[qi] = e.target.value;
+                          return next;
+                        });
+                      }}
+                      placeholder="Type a custom response..."
+                      className={`w-full px-4 py-2 rounded-lg border border-dashed text-sm transition-colors bg-transparent placeholder:text-muted-foreground focus:outline-none ${
+                        isOtherSelected
+                          ? "border-primary bg-primary/10 text-foreground"
+                          : "border-border text-muted-foreground hover:border-primary/40"
+                      }`}
+                    />
+                  </>
+                ) : (
+                  <input
+                    type="text"
+                    value={customTexts[qi]}
+                    onFocus={() => {
+                      setSelections((prev) => {
+                        const next = [...prev];
+                        next[qi] = 0;
+                        return next;
+                      });
+                    }}
+                    onChange={(e) => {
+                      setSelections((prev) => {
+                        const next = [...prev];
+                        next[qi] = 0;
+                        return next;
+                      });
+                      setCustomTexts((prev) => {
+                        const next = [...prev];
+                        next[qi] = e.target.value;
+                        return next;
+                      });
+                    }}
+                    placeholder="Type your answer..."
+                    className="w-full px-4 py-2 rounded-lg border text-sm transition-colors bg-transparent placeholder:text-muted-foreground focus:outline-none border-border text-foreground hover:border-primary/40 focus:border-primary"
+                  />
+                )}
+              </div>
+            );
+          })}
+        </div>
+
+        {/* Submit */}
+        <div className="px-5 pb-4">
+          <button
+            onClick={handleSubmit}
+            disabled={!canSubmit}
+            className="w-full flex items-center justify-center gap-2 py-2.5 rounded-lg text-sm font-medium bg-primary text-primary-foreground hover:bg-primary/90 disabled:opacity-30 disabled:cursor-not-allowed transition-colors"
+          >
+            <Send className="w-3.5 h-3.5" />
+            Submit All
+          </button>
+        </div>
+      </div>
+    </div>
+  );
+}

core/frontend/src/components/NodeDetailPanel.tsx

@@ -1,6 +1,6 @@
 import { useState, useEffect, useRef } from "react";
 import { X, Cpu, Zap, Clock, RotateCcw, CheckCircle2, AlertCircle, Loader2, ChevronDown, ChevronRight, Copy, Check, Terminal, Wrench, BookOpen, GitBranch, Bot } from "lucide-react";
-import type { GraphNode, NodeStatus } from "./AgentGraph";
+import type { GraphNode, NodeStatus } from "./graph-types";
 import type { NodeSpec, ToolInfo, NodeCriteria } from "../api/types";
 import { graphsApi } from "../api/graphs";
 import { logsApi } from "../api/logs";
@@ -299,13 +299,13 @@ function SubagentsTab({ subAgentIds, allNodeSpecs, subagentReports }: { subAgent
   );
 }
 
-type Tab = "overview" | "tools" | "logs" | "prompt" | "subagents";
+type Tab = "overview" | "breakdown" | "tools" | "logs" | "subagents";
 
 const tabs: { id: Tab; label: string; Icon: React.FC<{ className?: string }> }[] = [
   { id: "overview", label: "Overview", Icon: ({ className }) => <GitBranch className={className} /> },
+  { id: "breakdown", label: "Breakdown", Icon: ({ className }) => <BookOpen className={className} /> },
   { id: "tools", label: "Tools", Icon: ({ className }) => <Wrench className={className} /> },
   { id: "logs", label: "Logs", Icon: ({ className }) => <Terminal className={className} /> },
-  { id: "prompt", label: "Prompt", Icon: ({ className }) => <BookOpen className={className} /> },
   { id: "subagents", label: "Subagents", Icon: ({ className }) => <Bot className={className} /> },
 ];
 
@@ -331,7 +331,7 @@ export default function NodeDetailPanel({ node, nodeSpec, allNodeSpecs, subagent
 
   // Fetch real criteria when Overview tab is active and session is loaded
   useEffect(() => {
-    if (activeTab === "overview" && sessionId && graphId && node) {
+    if (activeTab === "breakdown" && sessionId && graphId && node) {
       graphsApi.nodeCriteria(sessionId, graphId, node.id, workerSessionId || undefined)
         .then(r => setRealCriteria(r))
         .catch(() => setRealCriteria(null));
@@ -410,6 +410,10 @@ export default function NodeDetailPanel({ node, nodeSpec, allNodeSpecs, subagent
       {/* Tab content */}
       <div className="flex-1 overflow-auto px-4 py-4 flex flex-col gap-3">
         {activeTab === "overview" && (
+          <SystemPromptTab systemPrompt={nodeSpec?.system_prompt} />
+        )}
+
+        {activeTab === "breakdown" && (
           <>
             <p className="text-[10px] font-medium text-muted-foreground uppercase tracking-wider">Action Plan</p>
             {actionPlan ? (
@@ -489,10 +493,6 @@ export default function NodeDetailPanel({ node, nodeSpec, allNodeSpecs, subagent
           <LogsTab nodeId={node.id} isActive={isActive} sessionId={sessionId} graphId={graphId} workerSessionId={workerSessionId} nodeLogs={nodeLogs} />
         )}
 
-        {activeTab === "prompt" && (
-          <SystemPromptTab systemPrompt={nodeSpec?.system_prompt} />
-        )}
-
         {activeTab === "subagents" && nodeSpec?.sub_agents && (
           <SubagentsTab
             subAgentIds={nodeSpec.sub_agents}

core/frontend/src/components/RunButton.tsx

@@ -0,0 +1,40 @@
+import { memo, useState } from "react";
+import { Play, Pause, Loader2, CheckCircle2 } from "lucide-react";
+import type { RunButtonProps } from "./graph-types";
+
+export const RunButton = memo(function RunButton({ runState, disabled, onRun, onPause, btnRef }: RunButtonProps) {
+  const [hovered, setHovered] = useState(false);
+  const showPause = runState === "running" && hovered;
+
+  return (
+    <button
+      ref={btnRef}
+      onClick={runState === "running" ? onPause : onRun}
+      disabled={runState === "deploying" || disabled}
+      onMouseEnter={() => setHovered(true)}
+      onMouseLeave={() => setHovered(false)}
+      className={`flex items-center gap-1.5 px-2.5 py-1 rounded-md text-[11px] font-semibold transition-all duration-200 ${
+        showPause
+          ? "bg-amber-500/15 text-amber-400 border border-amber-500/40 hover:bg-amber-500/25 active:scale-95 cursor-pointer"
+          : runState === "running"
+          ? "bg-green-500/15 text-green-400 border border-green-500/30 cursor-pointer"
+          : runState === "deploying"
+          ? "bg-primary/10 text-primary border border-primary/20 cursor-default"
+          : disabled
+          ? "bg-muted/30 text-muted-foreground/40 border border-border/20 cursor-not-allowed"
+          : "bg-primary/10 text-primary border border-primary/20 hover:bg-primary/20 hover:border-primary/40 active:scale-95"
+      }`}
+    >
+      {runState === "deploying" ? (
+        <Loader2 className="w-3 h-3 animate-spin" />
+      ) : showPause ? (
+        <Pause className="w-3 h-3 fill-current" />
+      ) : runState === "running" ? (
+        <CheckCircle2 className="w-3 h-3" />
+      ) : (
+        <Play className="w-3 h-3 fill-current" />
+      )}
+      {runState === "deploying" ? "Deploying\u2026" : showPause ? "Pause" : runState === "running" ? "Running" : "Run"}
+    </button>
+  );
+});

core/frontend/src/components/TopBar.tsx

@@ -1,8 +1,8 @@
 import { useState, useCallback } from "react";
 import { useNavigate } from "react-router-dom";
 import { Crown, X } from "lucide-react";
-import { loadPersistedTabs, savePersistedTabs, TAB_STORAGE_KEY, type PersistedTabState } from "@/lib/tab-persistence";
 import { sessionsApi } from "@/api/sessions";
+import { loadPersistedTabs, savePersistedTabs, TAB_STORAGE_KEY, type PersistedTabState } from "@/lib/tab-persistence";
 
 export interface TopBarTab {
   agentType: string;
@@ -51,10 +51,10 @@ export default function TopBar({ tabs: tabsProp, onTabClick, onCloseTab, canClos
       onCloseTab(agentType);
       return;
     }
-    // Kill the backend session (queen/judge/worker) even outside workspace
+    // Kill the backend session (queen/worker) even outside workspace
     sessionsApi.list()
       .then(({ sessions }) => {
-        const match = sessions.find(s => s.agent_path === agentType);
+        const match = sessions.find(s => s.agent_path.endsWith(agentType));
         if (match) return sessionsApi.stop(match.session_id);
       })
       .catch(() => {});  // fire-and-forget

core/frontend/src/components/graph-types.ts

@@ -0,0 +1,28 @@
+export type NodeStatus = "running" | "complete" | "pending" | "error" | "looping";
+
+export type NodeType = "execution" | "trigger";
+
+export interface GraphNode {
+  id: string;
+  label: string;
+  status: NodeStatus;
+  nodeType?: NodeType;
+  triggerType?: string;
+  triggerConfig?: Record<string, unknown>;
+  next?: string[];
+  backEdges?: string[];
+  iterations?: number;
+  maxIterations?: number;
+  statusLabel?: string;
+  edgeLabels?: Record<string, string>;
+}
+
+export type RunState = "idle" | "deploying" | "running";
+
+export interface RunButtonProps {
+  runState: RunState;
+  disabled: boolean;
+  onRun: () => void;
+  onPause: () => void;
+  btnRef: React.Ref<HTMLButtonElement>;
+}

core/frontend/src/index.css

@@ -72,6 +72,33 @@
     --border: 240 3.7% 15.9%;
     --input: 240 3.7% 15.9%;
     --ring: 45 93% 47%;
+
+    /* Agent graph node status colors */
+    --node-running: 45 95% 58%;
+    --node-looping: 38 90% 55%;
+    --node-complete: 43 70% 45%;
+    --node-pending: 35 15% 28%;
+    --node-pending-bg: 35 10% 12%;
+    --node-pending-border: 35 10% 20%;
+    --node-error: 0 65% 55%;
+
+    /* Agent graph trigger node colors */
+    --trigger-bg: 210 25% 14%;
+    --trigger-border: 210 30% 30%;
+    --trigger-text: 210 30% 65%;
+    --trigger-icon: 210 40% 55%;
+
+    /* Draft graph chrome colors */
+    --draft-edge: 220 10% 30%;
+    --draft-edge-arrow: 220 10% 35%;
+    --draft-edge-label: 220 10% 45%;
+    --draft-back-edge: 220 10% 25%;
+    --draft-group-fill: 220 15% 18%;
+    --draft-group-stroke: 220 10% 40%;
+    --draft-chrome-text: 220 10% 50%;
+    --draft-chrome-text-dim: 220 10% 55%;
+    --draft-node-text: 0 0% 78%;
+    --draft-node-text-hover: 0 0% 92%;
   }
 }
 

core/frontend/src/lib/chat-helpers.test.ts

@@ -1,60 +1,6 @@
 import { describe, it, expect } from "vitest";
-import { backendMessageToChatMessage, sseEventToChatMessage, formatAgentDisplayName } from "./chat-helpers";
-import type { AgentEvent, Message } from "@/api/types";
-
-// ---------------------------------------------------------------------------
-// backendMessageToChatMessage
-// ---------------------------------------------------------------------------
-
-describe("backendMessageToChatMessage", () => {
-  it("converts a user message", () => {
-    const msg: Message = { seq: 1, role: "user", content: "hello", _node_id: "chat" };
-    const result = backendMessageToChatMessage(msg, "inbox-management");
-    expect(result.type).toBe("user");
-    expect(result.agent).toBe("You");
-    expect(result.role).toBeUndefined();
-    expect(result.content).toBe("hello");
-    expect(result.thread).toBe("inbox-management");
-  });
-
-  it("converts an assistant message with node_id as agent", () => {
-    const msg: Message = { seq: 2, role: "assistant", content: "hi", _node_id: "intake" };
-    const result = backendMessageToChatMessage(msg, "inbox-management");
-    expect(result.agent).toBe("intake");
-    expect(result.role).toBe("worker");
-    expect(result.type).toBeUndefined();
-  });
-
-  it("defaults agent to 'Agent' when _node_id is empty", () => {
-    const msg: Message = { seq: 3, role: "assistant", content: "ok", _node_id: "" };
-    const result = backendMessageToChatMessage(msg, "inbox-management");
-    expect(result.agent).toBe("Agent");
-  });
-
-  it("produces deterministic ID from seq", () => {
-    const msg: Message = { seq: 42, role: "user", content: "test", _node_id: "x" };
-    const result = backendMessageToChatMessage(msg, "thread");
-    expect(result.id).toBe("backend-42");
-  });
-
-  it("passes through the thread parameter", () => {
-    const msg: Message = { seq: 1, role: "user", content: "hi", _node_id: "x" };
-    const result = backendMessageToChatMessage(msg, "my-thread");
-    expect(result.thread).toBe("my-thread");
-  });
-
-  it("uses agentDisplayName instead of node_id when provided", () => {
-    const msg: Message = { seq: 2, role: "assistant", content: "hi", _node_id: "intake" };
-    const result = backendMessageToChatMessage(msg, "thread", "Competitive Intel Agent");
-    expect(result.agent).toBe("Competitive Intel Agent");
-  });
-
-  it("still shows 'You' for user messages even when agentDisplayName is provided", () => {
-    const msg: Message = { seq: 1, role: "user", content: "hello", _node_id: "chat" };
-    const result = backendMessageToChatMessage(msg, "thread", "My Agent");
-    expect(result.agent).toBe("You");
-  });
-});
+import { sseEventToChatMessage, formatAgentDisplayName } from "./chat-helpers";
+import type { AgentEvent } from "@/api/types";
 
 // ---------------------------------------------------------------------------
 // sseEventToChatMessage
@@ -250,6 +196,102 @@ describe("sseEventToChatMessage", () => {
     );
   });
 
+  it("different inner_turn values produce different message IDs", () => {
+    const e1 = makeEvent({
+      type: "client_output_delta",
+      node_id: "queen",
+      execution_id: "exec-1",
+      data: { snapshot: "first response", iteration: 0, inner_turn: 0 },
+    });
+    const e2 = makeEvent({
+      type: "client_output_delta",
+      node_id: "queen",
+      execution_id: "exec-1",
+      data: { snapshot: "after tool call", iteration: 0, inner_turn: 1 },
+    });
+    const r1 = sseEventToChatMessage(e1, "t");
+    const r2 = sseEventToChatMessage(e2, "t");
+    expect(r1!.id).not.toBe(r2!.id);
+  });
+
+  it("same inner_turn produces same ID (streaming upsert within one LLM call)", () => {
+    const e1 = makeEvent({
+      type: "client_output_delta",
+      node_id: "queen",
+      execution_id: "exec-1",
+      data: { snapshot: "partial", iteration: 0, inner_turn: 1 },
+    });
+    const e2 = makeEvent({
+      type: "client_output_delta",
+      node_id: "queen",
+      execution_id: "exec-1",
+      data: { snapshot: "partial response", iteration: 0, inner_turn: 1 },
+    });
+    expect(sseEventToChatMessage(e1, "t")!.id).toBe(
+      sseEventToChatMessage(e2, "t")!.id,
+    );
+  });
+
+  it("absent inner_turn produces same ID as inner_turn=0 (backward compat)", () => {
+    const withField = makeEvent({
+      type: "client_output_delta",
+      node_id: "queen",
+      execution_id: "exec-1",
+      data: { snapshot: "hello", iteration: 2, inner_turn: 0 },
+    });
+    const withoutField = makeEvent({
+      type: "client_output_delta",
+      node_id: "queen",
+      execution_id: "exec-1",
+      data: { snapshot: "hello", iteration: 2 },
+    });
+    expect(sseEventToChatMessage(withField, "t")!.id).toBe(
+      sseEventToChatMessage(withoutField, "t")!.id,
+    );
+  });
+
+  it("inner_turn=0 produces no suffix (matches old ID format)", () => {
+    const event = makeEvent({
+      type: "client_output_delta",
+      node_id: "queen",
+      execution_id: "exec-1",
+      data: { snapshot: "hello", iteration: 3, inner_turn: 0 },
+    });
+    const result = sseEventToChatMessage(event, "t");
+    expect(result!.id).toBe("stream-exec-1-3-queen");
+  });
+
+  it("inner_turn>0 adds -t suffix to ID", () => {
+    const event = makeEvent({
+      type: "client_output_delta",
+      node_id: "queen",
+      execution_id: "exec-1",
+      data: { snapshot: "hello", iteration: 3, inner_turn: 2 },
+    });
+    const result = sseEventToChatMessage(event, "t");
+    expect(result!.id).toBe("stream-exec-1-3-t2-queen");
+  });
+
+  it("llm_text_delta also uses inner_turn for distinct IDs", () => {
+    const e1 = makeEvent({
+      type: "llm_text_delta",
+      node_id: "research",
+      execution_id: "exec-1",
+      data: { snapshot: "first", inner_turn: 0 },
+    });
+    const e2 = makeEvent({
+      type: "llm_text_delta",
+      node_id: "research",
+      execution_id: "exec-1",
+      data: { snapshot: "second", inner_turn: 1 },
+    });
+    const r1 = sseEventToChatMessage(e1, "t");
+    const r2 = sseEventToChatMessage(e2, "t");
+    expect(r1!.id).not.toBe(r2!.id);
+    expect(r1!.id).toBe("stream-exec-1-research");
+    expect(r2!.id).toBe("stream-exec-1-t1-research");
+  });
+
   it("uses timestamp fallback when both turnId and execution_id are null", () => {
     const event = makeEvent({
       type: "client_output_delta",
@@ -261,25 +303,36 @@ describe("sseEventToChatMessage", () => {
     expect(result!.id).toMatch(/^stream-t-\d+-chat$/);
   });
 
-  it("converts client_input_requested with prompt to message", () => {
+  it("returns null for client_input_requested (handled in workspace.tsx)", () => {
     const event = makeEvent({
       type: "client_input_requested",
       node_id: "chat",
       execution_id: "abc",
       data: { prompt: "What next?" },
     });
-    const result = sseEventToChatMessage(event, "t");
-    expect(result).not.toBeNull();
-    expect(result!.content).toBe("What next?");
-    expect(result!.role).toBe("worker");
+    expect(sseEventToChatMessage(event, "t")).toBeNull();
   });
 
-  it("returns null for client_input_requested without prompt", () => {
+  it("converts client_input_received to user message", () => {
     const event = makeEvent({
-      type: "client_input_requested",
-      node_id: "chat",
+      type: "client_input_received",
+      node_id: "queen",
       execution_id: "abc",
-      data: { prompt: "" },
+      data: { content: "do the thing" },
+    });
+    const result = sseEventToChatMessage(event, "t");
+    expect(result).not.toBeNull();
+    expect(result!.agent).toBe("You");
+    expect(result!.type).toBe("user");
+    expect(result!.content).toBe("do the thing");
+  });
+
+  it("returns null for client_input_received with empty content", () => {
+    const event = makeEvent({
+      type: "client_input_received",
+      node_id: "queen",
+      execution_id: "abc",
+      data: { content: "" },
     });
     expect(sseEventToChatMessage(event, "t")).toBeNull();
   });