pr-triage

Fetch PR comment metadata first and expand bodies on demand.

729|67|Updated Dec 8, 2025
One-click install
npx skills add https://github.com/joelhooks/swarm-tools --skill pr-triage
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pr-triage
Source: https://github.com/joelhooks/swarm-tools/tree/main/.opencode/skill/pr-triage
Command: npx skills add https://github.com/joelhooks/swarm-tools --skill pr-triage

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires zod, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Fetches PR comment metadata first and loads full bodies only when needed to reduce context and payloads.

Core Features & Use Cases

  • Metadata-first triage: fetch lightweight summaries for all comments to minimize data loaded into context.
  • On-demand body hydration: retrieve full comment bodies only for actionable items.
  • Automated decision workflow: categorize by severity, propose fixes with code blocks when needed, and decide to reply or resolve.
  • Use Case: In a large PR with dozens of comments, triage completes in minutes by focusing on metadata, then expanding only the critical threads and applying fixes with templates.

Quick Start

Run the triage workflow from the command line: bun run pr-comments.ts triage owner/repo 42 Or fetch metadata only: bun run pr-comments.ts list owner/repo 42 Then expand a comment: bun run pr-comments.ts expand owner/repo 123456

Frequently Asked Questions about pr-triage

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

FAQPage Schema
How do I triage GitHub PR comments without loading too much context?

PR comment triage minimizes context by fetching lightweight metadata first and expanding full bodies only for actionable items. This metadata-first approach drastically reduces payload size when reviewing large GitHub pull requests.

How do I categorize GitHub PR comments by severity during code review?

Automated PR triage workflows categorize comments by severity and category, providing structured guidance with templates for replies or resolutions. This allows you to quickly propose fixes with code blocks and decide whether to reply or resolve.

What is the best way to review large GitHub PRs with dozens of comments?

For large GitHub PRs, a metadata-first fetch retrieves lightweight summaries for all comments to minimize loaded data. You can then expand only critical threads on-demand to apply fixes using templates, completing triage in minutes.

Does this PR triage workflow integrate with GitHub REST and GraphQL APIs?

Yes, the PR comment triage workflow integrates with GitHub REST and GraphQL APIs for streamlined data fetching. This integration supports the metadata-first fetch and on-demand body expansion required for efficient automated review workflows.

How do I fetch PR comment metadata only without the full text?

You can fetch PR comment metadata only by running the list command to retrieve lightweight summaries. This avoids loading full comment bodies into context until you specifically expand a selected comment thread on-demand.

Can I automate GitHub code review decisions for actionable comments?

Automated decision workflows enable you to categorize PR comments by severity, propose fixes with code blocks when needed, and decide to reply or resolve. This structured triage process is essential for efficient automated code reviews.