github-triage

Analyze open GitHub issues and PRs into evidence-backed triage reports.

Updated Apr 14, 2026
One-click install
npx skills add https://github.com/gerald-ica/opencode-config-snapshot --skill github-triage-gerald-ica
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: github-triage
Source: https://github.com/gerald-ica/opencode-config-snapshot/tree/main/opencode/skills/omo-github-triage
Command: npx skills add https://github.com/gerald-ica/opencode-config-snapshot --skill github-triage-gerald-ica

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Analyzes all open GitHub issues and PRs and writes evidence-backed reports to /tmp/{datetime}/. Each claim includes a GitHub permalink as proof, and no actions are taken on GitHub.

Core Features & Use Cases

  • One item = one background quick task, producing an independent report for each issue or PR.
  • Every finding is supported by a permalink to the exact commit or file in GitHub.
  • ZERO mutation: no comments, merges, or label changes; reports are the only output.

Quick Start

Run the triage on a repository to generate per-item reports in the temporary directory.

Frequently Asked Questions about github-triage

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate read-only triage reports for open GitHub issues and PRs?

Read-only GitHub triage reports are generated by analyzing open issues and PRs, launching a dedicated background subagent for each item to produce an independent report with evidence-backed permalinks in a timestamped /tmp directory.

Does GitHub triage modify repository labels, comments, or pull request statuses?

GitHub triage performs zero mutations, meaning no comments, merges, or label changes are applied to the repository; the only output generated is the collection of read-only triage reports written to the /tmp directory.

Can I get exact line references and GitHub permalinks as proof in automated triage reports?

Automated triage reports include exact GitHub permalinks and line references as proof for all factual claims, ensuring every finding links directly to the specific commit or file within the repository for verification.

What is the best way to analyze a large backlog of open GitHub issues without making changes?

Analyzing a large backlog is handled by assigning one background quick task per issue or PR, producing independent read-only reports in a timestamped temporary directory without performing any mutations on the GitHub repository.

How do I run GitHub triage to write per-item reports to a temporary directory?

Running GitHub triage writes evidence-backed per-item reports to /tmp/{datetime}/, where each open issue and PR gets a dedicated background subagent that produces one independent report file with permalink citations.

Are there limitations to read-only GitHub triage regarding repository mutations?

Read-only GitHub triage is limited to analysis and reporting only, meaning it cannot perform repository mutations such as merging pull requests, posting comments, or modifying issue labels.