agentleFS
velonxvelonx/agent-skills/skills/github-integration/SKILL.md

github-integration

Skill1 starsChanged 48 days agoMITView on GitHub

Work with GitHub through the gh CLI - read and triage issues, summarise pull requests, check CI status, and draft issues or PR descriptions. Use when the user mentions GitHub issues, PRs, CI checks, releases, or asks for a repository activity summary.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

GitHub Integration

Overview

Uses the official gh CLI for common GitHub tasks so the agent works with the user's existing authentication and permissions — no tokens in prompts or files.

When to Use

  • "What's open on the repo?", "summarise PR #42", "why is CI failing?", "draft an issue for this bug"
  • Weekly repo activity summaries and triage.

Usage

Always read before you write. Anything that changes GitHub state (comment, label, close, merge, create) must be shown to the user and confirmed first.

  1. Check access: gh auth status. If not logged in, ask the user to run gh auth login themselves — never handle their token.
  2. Read with JSON output so results are parseable:
    • Issues: gh issue list --state open --limit 50 --json number,title,labels,updatedAt
    • PRs: gh pr list --json number,title,author,isDraft,reviewDecision
    • One PR: gh pr view 42 --json title,body,files,commits,reviews and gh pr diff 42
    • CI: gh pr checks 42, then gh run view <run-id> --log-failed
  3. Summarise: group issues by label/theme; for PRs give purpose, size, review state and blockers; for CI quote the failing step and the first real error line.
  4. Draft issues/PR bodies as Markdown: context, steps to reproduce or change summary, expected vs actual, checklist.
  5. Write only after confirmation, e.g. gh issue create --title "…" --body-file draft.md.

Examples

Prompt: "Why is CI red on my PR?" The agent runs gh pr checks, finds the failing job, pulls --log-failed, and reports: "test (node 20) fails in orders.test.ts:41 — expected 403, got 200", with the likely cause.

Requirements

  • gh installed and authenticated by the user.

Limitations

  • Limited to what the user's gh account can see and do.
  • Large logs are truncated to the failing step.

Changelog

  • 1.0.0 — Initial release.

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.