Install
$ agentstack add skill-euxx-claude-skills-for-copilot-gh-fix-ci ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
Fix Failing CI Checks
Use gh to locate failing PR checks, fetch GitHub Actions logs, summarize the failure, then propose a fix plan and implement only after explicit user approval.
Prerequisites
Verify gh authentication before doing anything else:
gh auth status
If unauthenticated, ask the user to run gh auth login (repo + workflow scopes required), then stop.
Workflow
1. Resolve the PR
# if the user specified a PR number or URL:
gh pr view --json number,url,headRefName
# otherwise (current branch):
gh pr view --json number,url,headRefName
Extract owner and repo from the returned url field — PR URLs have the form https://github.com///pull/, so split on / to get the 4th and 5th segments. These are the base (target) repository coordinates where GitHub Actions checks run, and are reliable even for cross-fork PRs.
2. Inspect failing checks
gh pr checks -R / --json name,state,bucket,link,startedAt,completedAt,workflow
If a field is rejected, rerun with the available fields reported by gh. Scope only to GitHub Actions checks. If link is not a GitHub Actions run URL, label it as external, report the URL only, and do not investigate further (e.g. Buildkite, CircleCI are out of scope).
3. Fetch failure logs
For each failing GitHub Actions check:
- Extract the run ID from
link(the numeric segment in the URL). - Fetch run metadata:
`` gh run view -R / --json name,workflowName,conclusion,status,url,event,headBranch,headSha ``
- Fetch logs:
`` gh run view -R / --log ``
- If the run is still in progress, get the job IDs and fetch logs directly:
`` gh run view -R / --json jobs ` Then for each relevant job: ` gh api "/repos///actions/jobs//logs" ``
Extract a concise failure snippet (relevant error lines and stack traces only).
4. Summarize failures
Present a short summary for the user:
- Check name and run URL
- Root cause (1–2 sentences with supporting log snippet)
- Note if logs are missing or truncated
5. Plan and get approval
Draft a concise fix plan covering:
- Which files to change and why
- Any tests that need updating
- Expected outcome
Ask the user for explicit approval before implementing.
6. Implement
Apply the approved plan. Show a summary of diffs and note any tests run.
7. Recheck
After changes, suggest:
gh pr checks -R / --watch
to confirm the checks pass.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: euxx
- Source: euxx/claude-skills-for-copilot
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.