pr-review-mining

Mine merged GitHub PR descriptions and review comments to extract contracts and architectural decisions.

8|1|Updated Jan 21, 2026
One-click install
npx skills add https://github.com/orban/intent-layer --skill pr-review-mining
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pr-review-mining
Source: https://github.com/orban/intent-layer/tree/main/skills/intent-layer/pr-review-mining
Command: npx skills add https://github.com/orban/intent-layer --skill pr-review-mining

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill extracts tribal knowledge from merged PR discussions to populate Intent Layer nodes, capturing rationale, contracts, and potential pitfalls that are buried in conversations.

Core Features & Use Cases

  • Extract rationale from PR descriptions (What/Why sections) to surface Architecture Decisions and Contracts.
  • Ingest reviewer feedback to identify Pitfalls and Anti-patterns, enabling proactive safeguards.
  • Populate AGENTS.md by aggregating findings per directory, ready for human review and decision logging.
  • Use Case: When a directory undergoes multiple PRs, summarize the key decisions and changes to guide future development.

Quick Start

Run the mining workflow against a target directory using the GitHub CLI to fetch merged PRs and comments and then generate structured findings.

Frequently Asked Questions about pr-review-mining

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

FAQPage Schema
How do I extract architecture decisions and pitfalls from merged GitHub PRs?

You can extract architecture decisions and pitfalls from merged GitHub PRs by mining PR descriptions and review comments to surface rationale and anti-patterns. This skill aggregates reviewer feedback and What/Why sections to identify potential pitfalls and document decisions for future development.

Can I use GitHub CLI to fetch PR review comments for analysis?

Yes, GitHub CLI can fetch PR review comments for analysis. This skill requires access to PR data via GitHub CLI or API to retrieve merged PR descriptions and comments, applying keyword-driven extraction with confidence scoring to surface contracts and architectural decisions.

What is the best way to document tribal knowledge from PR discussions in AGENTS.md?

Documenting tribal knowledge from PR discussions in AGENTS.md involves aggregating findings per directory touched by PRs. This skill populates AGENTS.md with structured findings extracted from PR descriptions and reviewer feedback, capturing rationale, contracts, and pitfalls ready for human review.

How do I mine merged PR descriptions to populate Intent Layer nodes with contracts?

To mine merged PR descriptions and populate Intent Layer nodes with contracts, apply keyword-driven extraction to PR data. This skill surfaces What/Why sections from PR descriptions to extract contracts and architectural decisions, then aggregates these structured findings per directory for Intent Layer population.

Does pr-review-mining support confidence scoring when extracting insights from PR data?

Yes, pr-review-mining supports confidence scoring when extracting insights from PR data. The skill applies keyword-driven extraction with confidence scoring to merged GitHub PR descriptions and review comments, ensuring structured findings like pitfalls and architectural decisions are reliably surfaced for AGENTS.md.

How do I summarize key decisions and changes when a directory undergoes multiple PRs?

Summarize key decisions and changes for a directory undergoing multiple PRs by mining merged PR descriptions and review comments. This skill aggregates reviewer feedback and rationale per directory, extracting architectural decisions and anti-patterns to guide future development and populate AGENTS.md.