b5aadcd51e
feat: auto-track popup pages and improve session startup logging
2026-03-12 16:49:46 -07:00
290d2f6823
feat: add --no-startup-window to Chrome launch flags
2026-03-12 16:49:36 -07:00
9f3339650d
chore: linter update
2026-03-12 14:27:17 -07:00
d5e5d3e83d
feat: add subagent activity tracking to queen status and instructions
2026-03-12 14:26:49 -07:00
5ea27dda09
refactor: update GCU system prompt for auto-snapshots and batching
2026-03-12 14:26:38 -07:00
6f9066ef20
feat: return auto-snapshot from browser interaction tools
2026-03-12 14:26:24 -07:00
c37185732a
feat: kill orphaned Chrome processes on GCU server shutdown
2026-03-12 14:26:05 -07:00
0c900fb50e
refactor: clean session startup and add page lifecycle management
2026-03-12 14:25:16 -07:00
4d3ac28878
feat: launch Chrome on macOS via open -n to coexist with user's browser
2026-03-12 14:24:55 -07:00
270c1f8c50
fix: use lazy %-formatting in subagent completion log to avoid f-string in logger
2026-03-12 14:24:30 -07:00
3d0859d06a
fix: stop clearing credentials_required on modal close to prevent infinite loop
2026-03-12 14:24:14 -07:00
ffe47c0f71
fix: credential modal eating errors, banner stays open
2026-03-12 09:41:53 -07:00
bf4652db4b
fix: share event bus so tool events are visible to parent
2026-03-12 08:41:34 -07:00
2acd526b71
feat: dynamic viewport sizing and suppress Chrome warning bar
2026-03-12 08:40:49 -07:00
df71834e4b
refactor: switch from Playwright browser to system Chrome via CDP
2026-03-12 08:39:43 -07:00
d87dfca1ab
Merge pull request #6075 from aden-hive/fix/credential-function-alignment
...
fix: align the credential functions to be the same
2026-03-11 15:11:57 -07:00
a79d7de482
[bug:6117:docs]: fix inconsistent configuration and troubleshooting guidance ( #6118 )
2026-03-11 14:41:54 +08:00
e5e57302fa
docs: fix typos and awkward copy ( #6115 )
...
* [bug:6109:README]: fix typos and awkward copy
* trigger ci
* rerun checks
2026-03-11 14:38:37 +08:00
c69cf1aea5
test(security): add comprehensive unit tests for 7 security scanning tools ( #6151 )
...
* test(security): add comprehensive unit tests for 7 security scanning tools
Add dedicated test files for all security scanning tools:
- test_dns_security_scanner.py (12 tests)
- test_http_headers_scanner.py (13 tests)
- test_ssl_tls_scanner.py (14 tests)
- test_subdomain_enumerator.py (15 tests)
- test_port_scanner.py (17 tests)
- test_tech_stack_detector.py (20 tests)
- test_risk_scorer.py (24 tests)
Total: 115 new tests covering:
- Input validation and cleaning
- Connection error handling
- Core scanning logic with mocked responses
- Grade/risk calculation
- Edge cases
Fixes #5920
* fix(tests): strengthen weak assertions in security scanner tests
- SSL scanner: replace always-true `or` assertions with specific checks
that verify hostname stripping actually happened
- Port scanner: verify timeout clamp value, not just absence of error
- DNS scanner: remove unused helper method
---------
Co-authored-by: hundao <alchemy_wimp@hotmail.com >
2026-03-11 13:29:11 +08:00
2f4cd8c36f
fix(credentials): improve exception handling in key_storage.py ( #6153 )
...
Replace bare except Exception: clauses with specific exception handling:
- delete_aden_api_key(): Catch FileNotFoundError, PermissionError at debug
level; log unexpected errors at WARNING with exc_info=True
- _read_credential_key_file(): Catch FileNotFoundError, PermissionError at
debug level; log unexpected errors at WARNING with exc_info=True
- _read_aden_from_encrypted_store(): Catch FileNotFoundError, PermissionError,
KeyError at debug level; log unexpected errors at WARNING with exc_info=True
This makes credential issues easier to diagnose by:
- Logging unexpected errors at WARNING level (visible in production)
- Including full stack traces with exc_info=True
- Keeping expected failures (file not found, permissions) at debug level
Fixes #5931
2026-03-11 13:05:10 +08:00
6f571e6d00
[BUG] fix: use ReplaceFileW for atomic writes on Windows to preserve ACLs ( #5849 )
...
* [BUG] fix: use ReplaceFileW for atomic writes on Windows to preserve ACLs
* fix: ensure atomic_replace checks for Windows API availability
2026-03-11 12:59:14 +08:00
31bc84106f
test: add API integration tests for hubspot, intercom, google_docs tools ( #6167 )
...
>>
>> Resolves #5921
>>
>> - test_hubspot_tool.py: 51 tests covering 15 MCP tools
>> - test_intercom_tool.py: 50 tests covering 11 MCP tools
>> - test_google_docs_tool.py: 57 tests covering 11 MCP tools
2026-03-11 12:55:03 +08:00
fd79dceb0f
Merge pull request #6166 from aden-hive/fix/subagent-reply-stall
...
Release / Create Release (push) Waiting to run
micro-fix: update escalation tests for new ESCALATION_REQUESTED flow
2026-03-10 19:47:00 -07:00
ad50139d67
chore: lint
2026-03-10 19:46:35 -07:00
12fb40c110
test: update escalation tests for ESCALATION_REQUESTED flow
...
Tests were asserting the old CLIENT_OUTPUT_DELTA + CLIENT_INPUT_REQUESTED
pattern; the fix in 89ccd66f
2026-03-10 19:45:21 -07:00
738e469d96
Merge pull request #6165 from aden-hive/feature/provider-moonshotai-kimi
...
feat: support MoonShot AI Kimi subscription
2026-03-10 19:39:25 -07:00
80ccbcc827
chore: lint
2026-03-10 19:37:18 -07:00
08fac31a9d
Merge pull request #6159 from aden-hive/fix/subagent-reply-stall
...
fix: route subagent report_to_parent escalations to queen instead of user
2026-03-10 18:24:33 -07:00
89ccd66fb9
fix: subagent _EscalationReceiver
2026-03-10 18:21:50 -07:00
7c47e367de
feat: support moonshotai kimi subscription
2026-03-10 18:03:44 -07:00
b8741bf94c
fix: queen agent system prompt hooks
2026-03-10 16:25:07 -07:00
c90dcbb32f
Merge pull request #6152 from aden-hive/refactor/remove-dead-code
...
refactor: remove deprecated codes
2026-03-10 15:31:34 -07:00
ac3a5f5e93
chore: remove the ai generated temp doc
2026-03-10 15:29:21 -07:00
1ccfdbbf7d
chore: minimax key check
2026-03-10 15:24:09 -07:00
1de37d2747
chore: lint
2026-03-10 15:00:14 -07:00
2aefdf5b5f
refactor: remove deprecated codes
2026-03-10 14:57:54 -07:00
4caaa79900
Merge pull request #5988 from roberthallers/docs/fix-tui-deprecation-5941
...
docs: fix TUI deprecation inconsistency in roadmap
2026-03-10 16:46:41 +08:00
296089d4cd
Merge pull request #6108 from Hundao/fix/subagent-judge-feedback
...
fix: SubagentJudge and implicit judge return feedback=None on ACCEPT
2026-03-10 15:39:29 +08:00
cae5f971cf
fix: update test assertions for newly added tools
...
Tool counts and expected lists were outdated after new tools were added
to stripe, linear, apollo, discord, and google_analytics.
2026-03-10 15:36:12 +08:00
bac716eea3
fix: pass feedback="" on evaluated ACCEPT verdicts in SubagentJudge and implicit judge
...
Fixes #6107
2026-03-10 15:24:39 +08:00
14daf672e8
Fix: SessionManager._cleanup_stale_active_sessions indiscriminately cancels healthy concurrent agent sessions ( #6081 )
...
* fixes a bug in the SessionManager
* chore: remove debug print from test
---------
Co-authored-by: hundao <alchemy_wimp@hotmail.com >
2026-03-10 15:18:11 +08:00
e352ae5145
fix(mcp): close errlog file handle to prevent resource leak ( #6094 )
...
Track the errlog file handle opened on non-Windows systems and
properly close it during cleanup to prevent file descriptor leaks.
Changes:
- Add _errlog_handle instance variable to track the file handle
- Store handle reference when opening os.devnull
- Close handle in _cleanup_stdio_async() after other cleanup
- Clear reference in disconnect() for safety
Fixes #6002
2026-03-10 15:06:51 +08:00
a58ffc2669
fix(server): use session.phase_state instead of session.mode_state in handle_pause ( #6069 )
...
The handle_pause endpoint referenced session.mode_state (lines 360-361),
which does not exist on the Session dataclass. This caused an
AttributeError every time the pause endpoint reached the phase transition
step, preventing the queen phase from transitioning to staging and
returning a 500 error to the frontend.
Changed to session.phase_state, consistent with handle_stop (line 412),
handle_run (line 75), and the Session dataclass definition
(session_manager.py line 44).
2026-03-10 15:03:19 +08:00
3fefea52be
Merge pull request #6102 from aden-hive/micro-fix/report-to-parent-empty-check
...
micro-fix: track reported_to_parent to prevent false empty-turn detection
2026-03-09 21:12:23 -07:00
06fd045b3e
micro-fix: track reported_to_parent to prevent false empty-turn detection
...
Turns that call report_to_parent were incorrectly treated as "truly
empty" because the flag was not propagated. Thread it through
_run_single_turn and include it in the empty-turn guard.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
2026-03-09 21:10:47 -07:00
2e43d2af46
Merge pull request #6100 from aden-hive/feature/integration-extended
...
Release / Create Release (push) Waiting to run
micro-fix: wrong reference for hive_coder
2026-03-09 19:52:35 -07:00
2c9790c65d
Merge remote-tracking branch 'origin' into feature/integration-extended
2026-03-09 19:52:17 -07:00
9700ac71bb
micro-fix: wrong reference for hive_coder
2026-03-09 19:50:07 -07:00
61ed67b068
Merge pull request #6097 from aden-hive/feature/integration-extended
...
Expand integration tool coverage across 40 vendors
2026-03-09 19:47:34 -07:00
c3bea8685a
Merge remote-tracking branch 'origin/main' into feature/integration-extended
2026-03-09 19:47:21 -07:00
bryan