One-click install
npx skills add https://github.com/anticorrelator/lore --skill pr-self-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pr-self-review
Source: https://github.com/anticorrelator/lore/tree/main/skills/pr-self-review
Command: npx skills add https://github.com/anticorrelator/lore --skill pr-self-review

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill helps developers conduct a thorough, self-reflective review of their own Pull Requests, catching potential issues and improving code quality before external review.

Core Features & Use Cases

  • Turn-based Dialog: Engages in a conversational review, exploring aspects of the PR one at a time.
  • Perspective Lenses: Utilizes different viewpoints (external reviewer, weakest assumption, invariant trace) to uncover insights.
  • Knowledge Integration: Enriches observations by querying a knowledge store for relevant context.
  • Work Item Generation: Creates actionable tasks for identified issues, distinguishing between implement-ready fixes and items needing further verification.
  • Use Case: After pushing a complex feature, use this Skill to simulate an external reviewer's perspective, ensuring all edge cases and architectural implications are considered.

Quick Start

Use the pr-self-review skill to review pull request number 42.

Frequently Asked Questions about pr-self-review

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

FAQPage Schema
How do I review my own pull request before requesting external review?

To review your own pull request, you can use a turn-based, reflective self-review process that simulates an external reviewer's perspective. This involves analyzing code diffs and identifying areas for exploration to improve code quality before external review.

What is the best way to catch edge cases in a complex GitHub pull request?

The best way to catch edge cases is to apply perspective lenses, such as the weakest assumption or invariant trace viewpoints. These lenses simulate an external reviewer's perspective to uncover insights and identify architectural implications in your code.

Can I simulate an external reviewer's perspective on my code diffs?

Yes, you can simulate an external reviewer's perspective to analyze your code diffs. The process uses a review protocol to explore aspects of the pull request one at a time, generating actionable work items for identified issues.

How do I generate actionable work items from a self-review of a pull request?

You generate actionable work items by synthesizing observations from your self-review and enriching them with knowledge store lookups. This process distinguishes between implement-ready fixes and items needing further verification for cross-boundary concerns.

Does the self-review process require the GitHub CLI to analyze pull requests?

Yes, analyzing pull requests through this self-review process requires the GitHub CLI (`gh`) dependency. It uses `gh` to facilitate the turn-based dialog and access the necessary pull request data for code diff analysis.