pr-learning

Mines GitHub PR review feedback into ranked, deduplicated Rule and Learning candidates with provenance tracking.

44|1|Updated Jan 21, 2026
One-click install
npx skills add https://github.com/cameroncooke/skills --skill pr-learning
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pr-learning
Source: https://github.com/cameroncooke/skills/tree/main/pr-learning
Command: npx skills add https://github.com/cameroncooke/skills --skill pr-learning

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires gh, python3, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill transforms raw PR review comments into durable team knowledge, preventing recurring mistakes and improving code quality over time.

Core Features & Use Cases

  • Extract Repeatable Patterns: Analyzes review comments to identify recurring issues or suggestions.
  • Propose Rules & Learnings: Generates ranked candidates for team-wide rules or best practices.
  • Codify Knowledge: Automatically adds approved insights to documentation (AGENTS.md/CLAUDE.md) with provenance.
  • Use Case: Automatically identify that multiple reviewers are flagging the same type of error in API error handling, then codify a new rule to prevent it in the future.

Quick Start

Use the pr-learning skill to analyze recent pull request feedback and suggest new rules or learnings.

Frequently Asked Questions about pr-learning

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

FAQPage Schema
How do I extract recurring rules from GitHub PR review comments?

To extract recurring rules from GitHub PR review comments, this Skill mines feedback across pull requests using the gh CLI, normalizes observations, and clusters repeated patterns to propose ranked strict Rules and soft Learnings with confidence scores.

What is the best way to turn pull request feedback into team documentation?

The best way to turn pull request feedback into team documentation is codifying approved insights directly into AGENTS.md or CLAUDE.md files, automatically tracking full provenance so recurring review suggestions become durable team knowledge.

Can I deduplicate new code review learnings against existing CLAUDE.md rules?

Yes, you can deduplicate new code review learnings against existing rules. The Skill compares extracted PR feedback observations against items already present in AGENTS.md and CLAUDE.md to prevent duplicate documentation entries.

Do I need the GitHub CLI to analyze pull request feedback for actionable insights?

Yes, you need the GitHub CLI (gh) installed. It is a required dependency used to collect and fetch PR review comments, alongside Python3 which executes the clustering and normalization logic for the analysis.

How does clustering code review feedback improve continuous improvement?

Clustering code review feedback improves continuous improvement by grouping normalized review observations into repeated patterns, allowing teams to identify systemic issues and codify them into strict Rules and soft Learnings rather than addressing isolated incidents.

What limitations exist when mining PR feedback for team-wide rules?

A limitation when mining PR feedback for team-wide rules is that the Skill relies entirely on the gh CLI and Python3 environment, meaning access permissions to the target repositories and proper local setup are required to fetch and process the review comments.