agentleFS
codervisorcodervisor/leanspec/.claude/skills/watch-ci/SKILL.md

watch-ci

Skill294 starsChanged 6 months agoMITView on GitHub

Watch GitHub Actions CI status for the current commit until completion. Use after pushing changes to monitor build results.

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If failed: which step failed and relevant error output

    line 80

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Watch CI

Poll the GitHub Actions CI pipeline for the current HEAD commit until all jobs finish.

Environment

The gh CLI is not available in the Claude VM. Use the GitHub REST API via curl. The repo is codervisor/leanspec.

Steps

  1. Get the current commit SHA and branch:
    SHA=$(git rev-parse HEAD)
    BRANCH=$(git branch --show-current)
    echo "Watching CI for commit $SHA on branch $BRANCH"
  2. Find the workflow run matching our exact commit. The API returns runs newest-first; filter by head_sha. If the run hasn't appeared yet (GitHub can take a few seconds), retry up to 5 times with 10s waits:
    curl -sH "Accept: application/vnd.github+json" \
      "https://api.github.com/repos/codervisor/lean-spec/actions/runs?branch=$BRANCH&head_sha=$SHA&per_page=1" \
      | python3 -c "
    import json, sys
    data = json.load(sys.stdin)
    runs = data.get('workflow_runs', [])
    if not runs:
        print('NO_RUNS'); sys.exit()
    r = runs[0]
    print(f'RUN_ID={r[\"id\"]}')
    print(f'Status: {r[\"status\"]}  Conclusion: {r.get(\"conclusion\") or \"pending\"}  Commit: {r[\"head_sha\"][:8]}')"
  3. Poll jobs until all complete (every 60s for Rust builds which take ~8-10 min):
    curl -sH "Accept: application/vnd.github+json" \
      "https://api.github.com/repos/codervisor/lean-spec/actions/runs/$RUN_ID/jobs" \
      | python3 -c "
    import json, sys
    data = json.load(sys.stdin)
    for j in data.get('jobs', []):
        steps = [s for s in j.get('steps', []) if s['status'] == 'in_progress']
        step_info = f' -> {steps[0][\"name\"]}' if steps else ''
        print(f'{j[\"name\"]:40}  {j[\"status\"]:12}  {j.get(\"conclusion\") or \"\"}{step_info}')"
  4. On failure, fetch the failed job logs to diagnose:
    curl -sH "Accept: application/vnd.github+json" \
      "https://api.github.com/repos/codervisor/lean-spec/actions/runs/$RUN_ID/jobs" \
      | python3 -c "
    import json, sys
    data = json.load(sys.stdin)
    for j in data.get('jobs', []):
        if j.get('conclusion') == 'failure':
            print(f'FAILED: {j[\"name\"]} (id: {j[\"id\"]})')
            for s in j.get('steps', []):
                if s.get('conclusion') == 'failure':
                    print(f'  Step: {s[\"name\"]}')"

CI Structure

  • node (~2 min): pnpm install, build, typecheck, test
  • rust (~8-10 min, depends on node): cargo fmt, clippy, build, test, TS binding export

Reporting

Give the user a brief status update each poll cycle. On completion, summarize:

  • Overall result (success/failure)
  • Per-job results
  • If failed: which step failed and relevant error output

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.