Track and rerun flaky GitHub Actions CI failures in Remotion issue #8375.
npx skills add https://github.com/danielbere1973/promoar --skill flake-danielbere1973
Please help me install this Agent Skill. Skill: flake Source: https://github.com/danielbere1973/promoar/tree/main/.agents/skills/flake Command: npx skills add https://github.com/danielbere1973/promoar --skill flake-danielbere1973
fetch failed error. The Skill confirms it is a known flake, increments its count in the tracker, and reruns the failed job. ## Quick Start Run the flake skill to investigate the failing CI check on my Remotion pull request, record it in the flake tracker, and rerun the failed job.High-intent search queries and answers about installing and using this skill.
Use the GitHub CLI to inspect failed workflow runs and job logs, extract a stable failure signature, and record it in a central tracker issue. This Skill maintains issue #8375 in the Remotion repository, incrementing counts for repeated signatures and appending new rows with evidence links.
Run `gh run rerun <run-id> --failed` to rerun only the failed jobs in a workflow run. For a single known flaky job in a completed run, use the API endpoint `gh api --method POST repos/<owner>/<repo>/actions/jobs/<job-id>/rerun`.
A failure is likely a flake if it passes on rerun without code changes, involves network or infrastructure errors like `fetch failed`, or is a timeout, race, or browser disconnect unrelated to the PR changes. Deterministic type errors, lint failures, and assertion failures from changed behavior are not flakes.
Yes. It lists recent open PRs with their check status and recent failed runs of the "Install and Test" workflow using `gh pr list` and `gh run list`, then drills into the relevant run and job logs.
A good signature includes the package or suite, the test name, and the root error text with timestamps, paths, ports, and ANSI codes stripped out. This groups repeats of the same flake while staying specific enough to act on.
volcengineManage GitHub pull requests, workflow runs, and API queries via gh CLI.
datadog-labsClassifies failing PR CI jobs as flaky, infra, or regression and proposes targeted remediation actions.
dylan-conlinTriage degraded orchestration environments with flight recording and spawn restrictions.
dxosInvestigate failing CI runs and quarantine flaky tests using Trunk's CI Autopilot MCP server.