2.6k
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
3 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill openclaw/skills --skill github-actions-retry-recovery-audit- _meta.json328 B
- SKILL.md2.1 KB
Overview
This skill audits GitHub Actions run exports to find fail-then-success retry sequences and quantify the minutes wasted before recovery. It surfaces the workflow/job slices with the largest cumulative rerun waste so teams can prioritize flaky targets. Use it to produce text or JSON reports and optionally fail CI when critical waste thresholds are exceeded.
How this skill works
The skill reads one or more GitHub Actions workflow run JSON files and groups attempts by repository, workflow, branch, and commit (headSha). It detects recovery sequences where one or more failure-like attempts are followed by a success, then sums elapsed minutes for failed attempts up to the first success. The output ranks recovery groups by wasted minutes and emits a summary, detailed recoveries, and a critical subset when thresholds are breached.
When to use it
- After collecting run JSON exports to identify where reruns repeatedly fail then later succeed.
- When you need to prioritize flaky workflows or jobs that waste CI minutes.
- As part of a CI health dashboard or periodic reliability audit.
- Before implementing retry policies or expensive root-cause investigations.
- To enforce fail gates that halt releases when critical rerun waste appears.
Best practices
- Collect run JSON consistently (same fields) so grouping by headSha and timestamps is reliable.
- Tune WARN and CRITICAL waste minute thresholds to match team SLA and CI cost targets.
- Filter by workflow, branch, or repository with the provided regex options to focus analysis.
- Use JSON output for automated dashboards and the text mode for quick human triage.
- Run the audit regularly and act on the highest-waste groups first to maximize savings.
Example use cases
- Generate a weekly report showing the top 20 recovery groups by wasted CI minutes.
- Fail a pipeline when any recovery group exceeds the critical waste threshold.
- Feed JSON output into an observability dashboard to track flaky rerun trends over time.
- Narrow investigation to a specific branch or workflow using WORKFLOW_MATCH and BRANCH_MATCH.
- Estimate potential CI cost savings by summing wasted minutes for targeted workflows.
FAQ
One or more GitHub Actions run JSON exports that include databaseId, workflowName, headBranch, headSha, conclusion, createdAt, updatedAt, url, and repository.
Can I automate failing my CI on bad results?
Yes. Set FAIL_ON_CRITICAL=1 and configure CRITICAL_WASTE_MINUTES; the skill exits non‑zero when critical recoveries are found.