Repository navigation
Closed
Labels
effort/sSmall: expected under 2 hours of workSmall: expected under 2 hours of workkind/bugSomething isn't working (prefixed form of the classic bug label)Something isn't working (prefixed form of the classic bug label)meta/enrichedIntake review done: acceptance criteria, links, and normalized labels addedIntake review done: acceptance criteria, links, and normalized labels addedpriority/highImportant and time-sensitiveImportant and time-sensitivesource/ci-watchFiled by the main-ci-failure-fix-trigger (fleet CI watch)Filed by the main-ci-failure-fix-trigger (fleet CI watch)task/triageQueue packet kind: verify, reproduce, and route this itemQueue packet kind: verify, reproduce, and route this item
Description
main CI failure — .github/workflows/agents.yml
- Run: https://github.com/developing-today/code/actions/runs/41567297665
- Workflow: .github/workflows/agents.yml
- Branch: main
- Head SHA: 4a29ec8
- Event: push
- Failed jobs: none — the run completed with zero jobs (it died at startup: invalid workflow file or platform failure). Triage first: confirm whether the workflow file is actually broken.
Failed log excerpt (gh run view 41567297665 --log-failed, first ~60 lines)
(no failed-job log: the run completed with zero jobs — startup failure. `gh run view --log-failed` returned: failed to get run log: log not found)
Filed by main-ci-failure-fix-trigger (fleet). One fix track per failing workflow; duplicates are suppressed until this issue closes.
Acceptance criteria
- A push to main produces an agents.yml run that starts its jobs (no zero-job startup failure).
- The invalid
timeout-minuteskey is gone from every reusable-workflow caller job in agents.yml. - Any remaining early job failure after startup is tracked as its own issue, not this one.
Metadata
Metadata
Assignees
Labels
effort/sSmall: expected under 2 hours of workSmall: expected under 2 hours of workkind/bugSomething isn't working (prefixed form of the classic bug label)Something isn't working (prefixed form of the classic bug label)meta/enrichedIntake review done: acceptance criteria, links, and normalized labels addedIntake review done: acceptance criteria, links, and normalized labels addedpriority/highImportant and time-sensitiveImportant and time-sensitivesource/ci-watchFiled by the main-ci-failure-fix-trigger (fleet CI watch)Filed by the main-ci-failure-fix-trigger (fleet CI watch)task/triageQueue packet kind: verify, reproduce, and route this itemQueue packet kind: verify, reproduce, and route this item
Activity
dezren39 commentedon Nov 3, 2026
🤖 Agent note · filed
dezren39 commentedon Nov 3, 2026
🤖 Agent note · picked up
Merge pull request #4746 from developing-today/queue/6731-4742-agents…
dezren39 commentedon Nov 3, 2026
Root-cause evidence (from new-issue-creator-code-b, this run — evidence only, not a pickup; this issue is claimed by lane code-i, task 6731):
.github/workflows/agents.yml:42setstimeout-minutes: 60on thescheduledjob, which is a reusable-workflow caller job (uses: developing-today/code/.github/workflows/pm-scheduled.yml@main, agents.yml:41). GitHub does not allowtimeout-minuteson a caller job — the workflow is invalid and fails at startup with zero jobs, exactly the symptom in run 41567297665.1d7d82743("ci: add job timeouts to agents.yml", 2026-11-03T02:32Z, Jules):git show 1d7d82743is a 1-line diff adding only that key. Main was green on this workflow before it.timeout-minutes: 60line from the caller job (a timeout, if wanted, belongs on jobs insidepm-scheduled.yml)..github/workflows/mcpx*.yml(mcpx.yml, citing mcpx testing: every formatter and linter in the CI matrix, pinned by the flake (and run the plugin's bun test) #1996), so no lint caught this; repo-wide actionlint findings are already tracked in actionlint: repo-wide findings outside the mcpx workflow set (jules concurrency.queue, SC2129, SC2016/SC2086 in run blocks) #2845 — no separate issue filed for that.36 remaining items
fix(pm-dispatch): give pm-dispatch-task other callers the .pm-central…
dezren39 commentedon Nov 4, 2026
🤖 Agent note · parked (CI pending)
.pm-centralcheckout) verified real and FIXED in ef12837 (pm-dispatch-smoke.yml + nix-ci.yml ci-fix job + pm-dispatch-ci-fix wrapper), dispositioned on the Codex thread (reply 4464917469)dezren39 commentedon Nov 4, 2026
🤖 Agent note · parked
github.tokennamed-value). The Codex P1 (comment 4464653392) was fixed in ef12837 and replied to (4464917469); no Codex re-review of ef12837 yet. I triggered a fresh verification dispatch of pm-scheduled from the PR branch: run 41798828490, queued at head ef12837.fix(pm-dispatch): drop github.token expression literal from pm-dispat…
dezren39 commentedon Nov 4, 2026
🤖 Agent note · parked
fix(pm-dispatch): load scheduled-agent actions from a .pm-central che…
dezren39 commentedon Nov 4, 2026
🤖 Agent note · done
.pm-centralcheckout pinned toref: main— i.e. main's unfixed manifest, not the PR branch's. The branch manifest was verified fixed by direct read (the${{ github.token }}literal in the pm-dispatch-task description is gone in 278bdf9), run 41798828490 already proved the main CI failure: .github/workflows/agents.yml (run 41567297665) #4742 action-download fix gets past the action checkout, PR checks were green, and the Codex P1 (comment 4464653392) was fixed in ef12837 and dispositioned on its thread (reply 4464917469). Merged under the standing merge criteria; post-merge scheduled runs from main are the end-to-end confirmation of the manifest fix.