Repository navigation
Closed as not planned
Labels
Description
Commit list:
- fix: repair CI contracts and regenerate workflow locks (fix: repair CI contracts and regenerate workflow locks #73925) (8085744)
- Mark four GitHub safe outputs as stable (Mark four GitHub safe outputs as stable #73916) (d3bef2d)
- fix: authorize resolved issue relationship repositories (fix: authorize resolved issue relationship repositories #73882) (644fcec)
- fix: isolate MCP script secrets per tool (fix: isolate MCP script secrets per tool #73887) (19002c9)
- fix: bound MCP identifier suggestion resource usage (fix: bound MCP identifier suggestion resource usage #73884) (a1f805b)
- Upgrade AWF to v0.29.5 and generate Docker enclave integration (Upgrade AWF to v0.29.5 and generate Docker enclave integration #73819) (b4a3f9b)
- audit: count only the audited attempt's evals invocations, once each, in the evals summary (audit: count only the audited attempt's evals invocations, once each, in the evals summary #73886) (96459bd)
- Linter Miner: use model routing (cost/economy) instead of pinning claude-sonnet-5.5 (Linter Miner: use model routing (cost/economy) instead of pinning claude-sonnet-5.5 #73839) (75c77dc)
- Linter Miner: have discussion-miner read GitHub through the
githubbridge instead ofgh(Linter Miner: have discussion-miner read GitHub through thegithubbridge instead ofgh#73871) (52d9aa1) - Add configurable three-day engine installation cooldown (Add configurable three-day engine installation cooldown #73829) (5fec33b)
- Record a not-evaluated evals marker instead of UNKNOWN answers when the judge never ran or failed (Record a not-evaluated evals marker instead of UNKNOWN answers when the judge never ran or failed #73704) (3e66bf4)
- fix: prevent context injection into prompt template directives (fix: prevent context injection into prompt template directives #73586) (8214416)
- Route Code Scanning Fixer with model routing (cost, balanced) (Route Code Scanning Fixer with model routing (cost, balanced) #73840) (4b2b27d)
- Route Agent Job Health Monitor with model routing (goal: cost, mode: economy) (Route Agent Job Health Monitor with model routing (goal: cost, mode: economy) #73033) (9545524)
- Preserve agent config permissions and restore only after PR checkout (Preserve agent config permissions and restore only after PR checkout #72940) (f8325d8)
- Route Contribution Check through model routing (cost/economy) (Route Contribution Check through model routing (cost/economy) #73837) (f4bfda4)
- [actions-aw-linter] Fix artifact execution detection for option-prefixed scripts ([actions-aw-linter] Fix artifact execution detection for option-prefixed scripts #73832) (1917f59)
- Route Matt Pocock Skills Reviewer with model routing instead of pinning claude-sonnet-5 (Route Matt Pocock Skills Reviewer with model routing instead of pinning claude-sonnet-5 #73836) (784bab3)
- Route Daily action/setup/* Security Audit with model routing instead of pinning opus (Route Daily action/setup/* Security Audit with model routing instead of pinning opus #73838) (58e6310)
- [jsweep] Clean work_queue_yaml.cjs ([jsweep] Clean work_queue_yaml.cjs #73699) (86ad365)
- fix(linters): unwrap parens in appendoneelement and bytesbufferstring argument matching (fix(linters): unwrap parens in appendoneelement and bytesbufferstring argument matching #73697) (443c9c3)
- [docs] Update Astro dependencies - 2026-11-01 ([docs] Update Astro dependencies - 2026-11-01 #73683) (afef7e2)
- [eslint-miner] Add no-dynamic-string-replacement rule ([eslint-miner] Add no-dynamic-string-replacement rule #73700) (f42abaf)
- test: verify MCP upload resource limits for alert 1038 (test: verify MCP upload resource limits for alert 1038 #73821) (812a8ca)
- test: cover alert 1039 asset link-following regressions (test: cover alert 1039 asset link-following regressions #73822) (345f7b0)
- [purelock] Lock down packageFunctionAliasInitializer, statusOnlyExpressionPath, graderJavaScriptReferenceName with pure-function test suite ([purelock] Lock down packageFunctionAliasInitializer, statusOnlyExpressionPath, graderJavaScriptReferenceName with pure-function test suite #73778) (9dca091)
- Share pinned TLA+ runtime and formal-methods instructions (Share TLA+/TLC tooling across agentic workflows #73816) (f9c9f37)
- fix(parser): isolate formatted import resolution errors (fix(parser): isolate formatted import resolution errors #73817) (4cd03aa)
- docs(instructions): document max-length on custom safe-output job inputs ([instructions] Sync instruction files with release v1.0.1 #73741) (f1f8bc6)
- chore: remove dead function AllowsWrite ([dead-code] chore: remove dead functions — 1 function removed #73785) (0ce31ff)
Summary:
Most recent activity is concentrated on CI/test reliability, MCP security and resource limits, and model-routing changes for task execution. There are also a number of targeted fixes in linting, parser behavior, and documentation, with several test additions to lock in regressions.
Bash tool test: ✅
Overall status: PASS
Run URL: https://github.com/github/gh-aw/actions/runs/41377779852
Timestamp: 2026-11-02T02:30:02.015+00:00
Generated by ⚡ Smoke Copilot MAI · copilot · mai11 · ~1 AIC · ⌖ ~1 AIC · ⊞ 20K · ◷
Comment /smoke-copilot-mai to run again
Add label smoke to run again
- expires on Nov 1, 2026, 8:32 PM UTC-08:00
Activity
github-actions commentedon Nov 2, 2026
This issue is being closed as outdated. A newer issue has been created: #73952
View newer issue
This action was performed automatically by the
Smoke Copilot MAIworkflow.