Context
From health check report #33 — Recommendation #6.
Problem
The health-check script (scripts/pr_review_health.sh) downloads failed run logs but does not distinguish between "run failed and log was fetched" vs "run failed and log could not be retrieved." When logs are unavailable, the report shows (log unavailable) but there is no ::warning:: annotation to alert the operator.
Proposed Fix
After downloading logs, check which failed runs produced only the (log unavailable) placeholder. Emit a ::warning:: for each missing log so the operator knows the diagnosis is incomplete.
Expected Impact
Operators will see explicit warnings when failure logs are missing, prompting manual investigation.
Urgency
LOW — informational improvement to observability.
Context
From health check report #33 — Recommendation #6.
Problem
The health-check script (
scripts/pr_review_health.sh) downloads failed run logs but does not distinguish between "run failed and log was fetched" vs "run failed and log could not be retrieved." When logs are unavailable, the report shows(log unavailable)but there is no::warning::annotation to alert the operator.Proposed Fix
After downloading logs, check which failed runs produced only the
(log unavailable)placeholder. Emit a::warning::for each missing log so the operator knows the diagnosis is incomplete.Expected Impact
Operators will see explicit warnings when failure logs are missing, prompting manual investigation.
Urgency
LOW — informational improvement to observability.