feat: implement issue #2139 — Retry cron should hold retries until a Claude window's real reset, read from the budget poller's published result (split 1 of 2 from the weekly-cap issue)

Merged
DO
1mo ago
+642-0

Problem

Retry cron should hold retries until a Claude window's real reset, read from the budget poller's published result (split 1 of 2 from the weekly-cap issue)

From the issue: Split 1 of 2 from #1863 (which timed out at dev-lead's maximum action budget and asked to be split). As the fleet maintainer, I want the retry cron to read the dry-run budget poller's published result to learn whether a Claude window is exhausted and when it really resets, so that a weekly cap suppresses re-dispatch until its reset instead of retrying every two hours for up to seven days.

Risk

Medium — changes GitHub Actions workflow behavior, which is exercised only post-merge; verify via the affected workflow runs.

Test plan

Tests added/updated: tests/dev-lead/unit/test_dev_lead_retry_budget_hold.bats. Verification: bash scripts/dev-lead-lint.sh (shellcheck --severity=warning) ran pre-commit; the bats suite runs in CI.

Rollback

Revert this PR. No non-revertible side effects (no tags, migrations, or external state).

Monitoring

Watch the affected workflow run(s) in the Actions tab and this PR's Lint check for regressions.

Closes #2139