Claude Skill MartBrowse skillsWhat is a Skill?
Back to list

GitHub Triage (Read-Only Analyzer)

Analyzes every open issue and PR in parallel and writes permalink-backed triage reports to /tmp — without ever mutating anything on GitHub.

Dev & CodingAdvanced67,5315,507AI score 8/10Last updated: Aug 9, 2026

What it does

  • Uses the gh CLI to fetch all open issues and PRs (with pagination) and classifies them into questions, bugs, feature requests, bugfix PRs, and other PRs.
  • Spawns one background subagent per item so the whole backlog is analyzed concurrently, each writing its own report.
  • Output lands in /tmp/{YYYYMMDD-HHmmss}/issue-123.md, pr-45.md, plus a rolled-up SUMMARY.md with counts and an "items requiring attention" list.
  • Two hard rules define the skill:
    1. Zero-action policy — no gh issue comment, gh pr merge, gh pr review, no POST/PUT/PATCH/DELETE. Reads and local file writes only.
    2. Evidence rule — every factual claim needs a commit-SHA permalink; unsupported statements must be marked [UNVERIFIED] or stated as "no evidence found". Branch-name links are rejected.
  • Bug reports get a verdict of CONFIRMED_BUG / NOT_A_BUG / ALREADY_FIXED / UNCLEAR, and for already-fixed cases it hunts down the fixing commit via git log / git blame.

Who it's for

  • Open-source maintainers buried under a large, stale issue backlog.
  • Team leads who want AI-assisted judgement but absolutely no AI writes to the repository.
  • Reviewers who want a single table of CI, review, mergeability, and risk across the PR queue.
  • Project managers capturing a backlog snapshot before a sprint or a handover.

Examples

  1. Bulk backlog cleanup — say "triage issues" and get a list of issues that are already fixed, each linked to the exact commit that fixed them, so you can close them with confidence.
  2. PR queue review — say "triage PRs" to receive per-PR merge-readiness tables with MERGE / REQUEST_CHANGES / NEEDS_REVIEW / WAIT recommendations that a human still has to act on.
  3. Draft answers for questions — for [Question] issues it greps the codebase and drafts an answer with permalinks to the relevant lines, ready for you to polish and post yourself.

· · · Install guide · · ·

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the .agents/skills/github-triage folder from the GitHub repo code-yeongyu/oh-my-openagent into my ~/.claude/skills/github-triage/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/code-yeongyu/oh-my-openagent.git /tmp/oh-my-openagent && mkdir -p ~/.claude/skills && cp -r /tmp/oh-my-openagent/.agents/skills/github-triage ~/.claude/skills/

This is a third-party skill. Check the source repository before installing.

  1. Prerequisites: install the GitHub CLI (gh --version to check, brew install gh on macOS) and authenticate with gh auth login. You also need jq (brew install jq).
  2. Clone the repo: git clone https://github.com/code-yeongyu/oh-my-openagent.git /tmp/oh-my-openagent
  3. Create the skills folder: mkdir -p ~/.claude/skills
  4. Copy the skill: cp -r /tmp/oh-my-openagent/.agents/skills/github-triage ~/.claude/skills/
  5. Verify: run ls ~/.claude/skills/github-triage/SKILL.md.
  6. Run it: cd into the repository you want analyzed, start Claude Code, and type "triage issues", "triage PRs", or "github triage".
  7. Read the output: open SUMMARY.md and the individual reports inside the /tmp/{timestamp}/ directory printed at the end of the run.

Note: the skill assumes a runtime with task_create and background subagents. Without those it may degrade to sequential execution or skip steps. On repos with hundreds of open items, test on a small repo first to gauge time and token cost.

View source on GitHubLicense: NOASSERTION