Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Add PR fix mode toggle docs + improved README.
What This Skill Does
Analyzes failing GitHub Actions runs to identify root causes and propose minimal fixes. Uses the GitHub CLI and API to fetch run metadata and logs, then produces a structured report with ranked causes and fix options. Supports an opt-in PR mode that automatically applies and opens a fix.
Replaces manual log-diving by fetching, parsing, and ranking failure causes in one step with a structured report.
When to use it
- Diagnosing why a CI pipeline broke after a merge
- Summarizing failed workflow logs without reading through thousands of lines
- Getting a minimal patch for a flaky test or misconfigured workflow step
- Checking which job and step caused a PR to go red
- Automating a fix PR for a known CI failure pattern
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Why did run 9823456789 fail on my-org/my-repo?
- 1Runs `gh run view 9823456789 --repo my-org/my-repo --json status,conclusion,headBranch,url,name` to get run metadata
- 2Runs `gh run view 9823456789 --repo my-org/my-repo --log-failed` to retrieve only failed step logs
- 3Parses logs to identify the failing job, step, and exact error excerpt
- 4Ranks likely root causes based on error patterns
- 5Returns a CI Autopsy report with failing steps, error excerpt, ranked causes, and fix options
Report showing that the `build` job failed at the `Install dependencies` step due to a lockfile mismatch, with a recommended fix to commit the updated lockfile
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.