trw-review-pr

Score code changes across six quality dimensions and output a standardized review report.

Updated Mar 25, 2026
One-click install
npx skills add https://github.com/wallter/trw-mcp --skill trw-review-pr
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: trw-review-pr
Source: https://github.com/wallter/trw-mcp/tree/main/src/trw_mcp/data/codex/skills/trw-review-pr
Command: npx skills add https://github.com/wallter/trw-mcp --skill trw-review-pr

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Reviews this option helps teams enforce a consistent, objective evaluation of code changes by applying the FRAMEWORK.md rubric to PRs or branches, producing a clear verdict and actionable improvements.

Core Features & Use Cases

  • Scores each quality dimension (Correctness, Tests, Security, Performance, Maintainability, Completeness) and returns an overall verdict.
  • Collects context from diffs, commit history, and related PRD references to inform the assessment.
  • Generates a standardized Code Review Report with weighted scores and recommended actions suitable for pre-merge decisions.
  • Use Case: in CI pipelines or manual reviews to validate changes before merging.

Quick Start

Provide a structured code review for a given PR or branch using the FRAMEWORK.md rubric.

Frequently Asked Questions about trw-review-pr

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

FAQPage Schema
How do I automate structured PR reviews against a rubric?

Automate structured PR reviews by applying a FRAMEWORK.md rubric to score correctness, tests, security, performance, maintainability, and completeness. This process collects diffs and commit history to generate a standardized report with weighted scores and a merge verdict.

What is the best way to enforce consistent code review scoring in CI pipelines?

Enforce consistent code review scoring by running a rubric-based evaluation in your CI pipeline that validates builds and assesses PR diffs. It outputs a standardized report with weighted scores across six quality dimensions and a final verdict for pre-merge decisions.

How does FRAMEWORK.md evaluate security and performance in a PR diff?

FRAMEWORK.md evaluates security and performance by collecting context from PR diffs, commit history, and related PRD references. It applies weighted scoring across multiple quality dimensions to produce actionable improvement recommendations and an overall verdict.

Does the code review process require a build verification check before generating a report?

Yes, the code review process requires build verification via trw_build_check before generating a report. This ensures the code changes are valid and executable prior to evaluating correctness, tests, security, and other quality dimensions.

Can I use this automated code review approach for local repository branches?

Yes, you can use this automated code review approach for local repository branches as well as GitHub PR workflows. It automatically collects diffs and history from branches to evaluate changes against the FRAMEWORK.md rubric and output a detailed review report.

What limitations exist when reviewing diffs with a standardized code review rubric?

A limitation of reviewing diffs with a standardized rubric is the dependency on build verification tools like trw_build_check and the presence of related PRD references. Without complete commit history or PRD context, the weighted scoring and generated verdict may lack full actionable context.