review-prs

Screen GitHub pull requests for security risks, CI status, and bot likelihood before commenting.

3|Updated Apr 13, 2026
One-click install
npx skills add https://github.com/Abhilash-003/mstack --skill review-prs-abhilash-003
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-prs
Source: https://github.com/Abhilash-003/mstack/tree/main/review
Command: npx skills add https://github.com/Abhilash-003/mstack --skill review-prs-abhilash-003

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It reduces maintainer time spent on open pull requests by running a fast, structured pre-review that highlights security, test coverage, breaking change risk, and potential bot behavior before any comments are posted.

Core Features & Use Cases

  • Security-first diff scanning: Detects hardcoded secret patterns, injection risks, new dependency additions, and permission/auth changes to determine PR risk.
  • Test coverage and breaking change checks: Flags when source changes appear without corresponding tests and identifies likely breaking changes from diff-level signals.
  • CI signal + bot suspicion triage: Summarizes GitHub check results and uses author/profile + PR heuristics (from detection rules) to gauge spam/bot likelihood.
  • Human-controlled posting: Generates draft review comments and requires explicit user approval for what gets posted.

Quick Start

Run pre-screening for open pull requests in your configured repo by asking: "review the open PRs".

Frequently Asked Questions about review-prs

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

FAQPage Schema
How do I pre-screen GitHub pull requests for security risks and test coverage?

Pre-screen GitHub pull requests by analyzing diffs for hardcoded secrets, injection risks, missing tests, and breaking changes. This structured pre-review highlights PR risk and test coverage gaps before posting comments, reducing manual maintainer triage time.

How does bot detection work for automated pull request review?

Bot detection for automated pull request review uses author profile data and PR heuristics to gauge spam or bot likelihood. It combines these behavioral signals with CI status checks to summarize PR quality and flag suspicious automated contributions.

Can I review specific pull request numbers instead of all open PRs?

Yes, you can review a specific pull request number instead of batching all open PRs. The tool pre-screens up to 50 open, non-draft, non-approved PRs by default, but optionally accepts a specific PR number for targeted triage.

Do I need GitHub CLI authentication to run CI checks and security scanning on PRs?

Yes, GitHub CLI availability and authentication are required to run CI checks and security scanning on PRs. The tool reads optional repo configuration from .mstack/config.yml to enable or disable security scans and test requirements.

What is the best way to prevent automated bots from posting unwanted pull request review comments?

To prevent unwanted pull request review comments, the tool generates draft comments and requires explicit user approval before posting anything. This human-controlled posting mechanism ensures only approved triage outcomes are published to GitHub.

What are the limitations of automated breaking change detection in pull request diffs?

Automated breaking change detection in pull request diffs is limited to identifying likely breaking changes from diff-level signals only. It reads diff content to flag risks but does not execute code, meaning runtime breaking behaviors may be missed.