pr-review

Review pull request diffs and generate tiered Critical/Important/Suggestion reports.

Updated Apr 6, 2026
One-click install
npx skills add https://github.com/lety-ai/lety-skill-hub --skill pr-review-lety-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pr-review
Source: https://github.com/lety-ai/lety-skill-hub/tree/main/plugins/pr-review/skills/pr-review
Command: npx skills add https://github.com/lety-ai/lety-skill-hub --skill pr-review-lety-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides fast, consistent, and high-confidence code reviews for pull requests in the Lety 2.0 monorepo by automatically classifying changed files, running narrow specialist reviewers in parallel, and producing a single prioritized action-oriented report so maintainers can merge safely with fewer review cycles.

Core Features & Use Cases

  • Automatic scope detection: Identifies backend, frontend, migration, infra, and test changes from a git diff and detects re-reviews by collecting previous findings.
  • Parallel specialist reviewers: Spawns independent backend and frontend subagents with strict confidence thresholds and rulebooks, and validates migrations inline.
  • Structured, tiered reporting: Synthesises subagent outputs into a single Critical / Important / Suggestion report plus an action plan and re-review status when applicable.
  • Use case: Use when a contributor says "review my changes" or opens a PR to develop so the skill can produce a precise, reproducible code review and re-review checklist.

Quick Start

Please review my PR branch feature/x against develop and produce a tiered Critical/Important/Suggestion report summarizing backend and frontend findings and an action plan.

Frequently Asked Questions about pr-review

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

FAQPage Schema
How do I automate code review for a NestJS and NextJS monorepo pull request?

Automating code review for a NestJS and NextJS monorepo pull request involves classifying backend, frontend, migration, and infra changes from a git diff. The system runs parallel subagents to synthesize a tiered Critical, Important, and Suggestion report with an action plan.

What is a tiered Critical/Important/Suggestion code review report?

A tiered Critical/Important/Suggestion code review report categorizes pull request findings by severity to help maintainers prioritize fixes. It synthesizes parallel backend and frontend reviewer outputs into a single structured document with an actionable merge plan.

How do I run a diff analysis on a PR branch against the develop branch?

To run a diff analysis on a PR branch against develop, provide the git diff or PR context. The system detects backend, frontend, migration, and test file changes, validates migrations, and generates a structured review with an action plan.

Can I use automated pull request reviews for re-reviewing previously fixed changes?

Automated pull request reviews support re-reviewing previously fixed changes by collecting previous findings. The system validates whether prior issues are resolved and updates the re-review status in the final structured report.

Does the code review skill validate TypeORM migrations automatically?

The code review skill validates TypeORM migrations automatically inline. When detecting migration changes within the monorepo diff, it checks for issues and synthesizes the results into the final tiered code review report.

What are the limitations of using parallel subagents for code review?

A limitation of using parallel subagents for code review is that they operate with strict confidence thresholds and predefined rulebooks. They are narrowly scoped to backend and frontend specialists, synthesizing only classified file types rather than providing general architectural feedback.