Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
Back to list

GitHub Triage (Read-Only)

Fans out one background agent per open issue/PR to produce evidence-backed, permalink-cited triage reports in /tmp — with zero writes to GitHub.

Dev & CodingAdvanced69,2955,711AI score 8/10Last updated: Sep 22, 2026

What it does

  • Uses the gh CLI to fetch every open issue and PR (with pagination past 500) and classifies them into questions, bugs, feature requests, bugfix PRs, and other.
  • Spawns one background quick subagent per item; each searches the codebase and writes its own report to /tmp/{YYYYMMDD-HHmmss}/issue-{N}.md or pr-{N}.md.
  • Every factual claim must carry a commit-SHA permalink. Anything unprovable is flagged [UNVERIFIED] or stated as "no evidence found". Branch-name links are rejected.
  • Bug reports get a verdict (CONFIRMED_BUG / NOT_A_BUG / ALREADY_FIXED / UNCLEAR) plus severity; ALREADY_FIXED requires locating the fixing commit via git log/git blame.
  • PRs get a merge-readiness table (CI, review, mergeable, risk) and a recommended action.
  • Ends with a SUMMARY.md roll-up of counts and items requiring attention.
  • Strict zero-action policy: no comments, merges, closes, labels, or any gh api -X POST/PUT/PATCH/DELETE.

Who it's for

  • Open-source maintainers digging out of an issue/PR backlog.
  • Engineers inheriting an unfamiliar repository who need a fast, cited status map.
  • Teams that want AI analysis but absolutely no AI-initiated writes to their repo.

Examples

  1. "triage issues" → 120 open issues classified; 14 identified as already fixed, each linked to the exact fixing commit.
  2. "triage PRs" → each bugfix PR verified against the root cause in the code, with a MERGE / REQUEST_CHANGES recommendation table.
  3. Pre-release check → skim only the "Items Requiring Attention" section of SUMMARY.md to find release blockers.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/HEAD/.agents/skills/github-triage/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

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 /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 GitHub CLI (gh) and run gh auth login. You also need git and jq.
  2. Clone the repository: git clone https://github.com/code-yeongyu/oh-my-openagent /tmp/oh-my-openagent
  3. Copy the skill into place: mkdir -p ~/.claude/skills && cp -r /tmp/oh-my-openagent/.agents/skills/github-triage ~/.claude/skills/
  4. cd into the repository you want to analyze and start Claude Code there.
  5. Type "github triage", "triage issues", or "triage PRs" to trigger the skill.
  6. Watch results stream in the terminal; full reports land in /tmp/{timestamp}/, with SUMMARY.md as the index.
  7. Note: the skill assumes a background subagent runtime (task, task_create, background_output). If your setup lacks it, ask the agent to process items sequentially instead.
View source on GitHubLicense: NOASSERTION