You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Root-cause the single failed run in the monitored window. Conclusion: this was a one-off wrong input to a manual run, not a workflow defect. No code change planned.
Findings
Window: 3 workflow_dispatch runs on main, 2026-10-21. One failed: run 39638248138.
The run was re-dispatched about 5 minutes later with discussion_number=6 and succeeded (39638933291). A later run, 39674859809, also succeeded.
scripts/initiative-planner/gather-context.sh is meant to fail loudly when an API call fails (see the header comment, "Any other API failure fails loud"). Stopping on a Discussion that doesn't exist is the correct behaviour, and the error message names the missing number.
The earlier history (2026-10-09) is all successes.
Implementation Checklist
None. Nothing in the code needs fixing, and adding special handling for a wrong number would weaken the fail-loud check on purpose-built input.
Tests to Write
None (no change in behaviour).
Test Command
bash .dev-lead/scripts/dev-lead-lint.sh (run to confirm the tree is unchanged and clean)
Recommendation: close as not-a-defect. The failure rate should drop below the threshold as later successful runs push this one out of the monitoring window.
β Auto-resolved by Fleet Monitor. This workflow's failure rate has been below the 10% threshold (or the workflow was removed) since 2026-10-22 (3 days) β the monitor stopped flagging it. Closing. It will reopen automatically if the workflow crosses the threshold again.
Activity
don-petry commentedon Oct 22, 2026
Dev-Lead Implementation Plan
Issue: #2389 β [Fleet Monitor] combsmith/.github-private β .github/workflows/initiative-planner.yml
Scope
Root-cause the single failed run in the monitored window. Conclusion: this was a one-off wrong input to a manual run, not a workflow defect. No code change planned.
Findings
workflow_dispatchruns onmain, 2026-10-21. One failed: run 39638248138.gather-context.sh:gh: Could not resolve to a Discussion with the number of 5.The run was dispatched withdiscussion_number=5. That Discussion does not exist in the target repo. The idea's Discussion there is feat: feature ideation framework β adversarial proposer/challenger/synthesizerΒ #6, titled "graduated from incubator feat: add Copilot engine support via REVIEW_ENGINE toggleΒ #5", so the incubator's number was entered by mistake.discussion_number=6and succeeded (39638933291). A later run, 39674859809, also succeeded.scripts/initiative-planner/gather-context.shis meant to fail loudly when an API call fails (see the header comment, "Any other API failure fails loud"). Stopping on a Discussion that doesn't exist is the correct behaviour, and the error message names the missing number.Implementation Checklist
Tests to Write
Test Command
bash .dev-lead/scripts/dev-lead-lint.sh(run to confirm the tree is unchanged and clean)Recommendation: close as not-a-defect. The failure rate should drop below the threshold as later successful runs push this one out of the monitoring window.
don-petry commentedon Oct 25, 2026
β Auto-resolved by Fleet Monitor. This workflow's failure rate has been below the 10% threshold (or the workflow was removed) since 2026-10-22 (3 days) β the monitor stopped flagging it. Closing. It will reopen automatically if the workflow crosses the threshold again.