review

Review the current branch against a base branch and output a severity-ranked report.

2|Updated Jan 22, 2026
One-click install
npx skills add https://github.com/abnegate/claudes --skill review-abnegate
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/abnegate/claudes/tree/main/skills/review
Command: npx skills add https://github.com/abnegate/claudes --skill review-abnegate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents broken, insecure, or inefficient changes from being merged by performing a thorough review of the current branch against the base branch with structured, parallelized analysis.

Core Features & Use Cases

  • Parallel, dimension-focused review: Runs security & data integrity, logic & correctness, performance & testing, and project standards checks in parallel to reduce review time while covering more ground.
  • Actionable severity-ranked output: Produces a consolidated report with deduplicated findings prioritized into Critical, Warnings, Suggestions, and Positive Notes.
  • Branch-aware context gathering: Automatically compares the current branch to an appropriate base branch (main vs a versioned base when a specific remote is present) and includes diffs and commit history for better traceability.

Use case: You have a feature branch ready for PR and want confidence that it won’t introduce vulnerabilities, regress behavior, degrade performance, or violate established code standards before merging.

Quick Start

Run the review skill to produce a severity-prioritized code review report comparing your current branch to the base branch.

Frequently Asked Questions about review

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

FAQPage Schema
How do I perform a thorough code review on a git branch before merging?

Perform a code review by comparing your current branch against a base branch. Gather commit context and diffs, run parallel checks for security and correctness, and output a severity-ranked report to catch critical issues before merging.

What is the best way to check a pull request for security vulnerabilities and performance regressions?

Checking a pull request for security and performance issues requires running parallel reviewer perspectives. This validates changes across security, data integrity, performance, and testing dimensions, deduplicating findings into a prioritized report of must-fix critical items.

Can I automate branch comparison to ensure changes meet engineering standards?

Automated branch comparison evaluates changes against engineering standards by analyzing the diff between your current and base branches. It automatically detects violations and outputs a consolidated report with warnings, suggestions, and positive notes to ensure compliance.

Does code review work with versioned base branches or only main?

Code review works with both main and versioned base branches. It automatically identifies the appropriate base by checking for a specific remote, comparing the current branch against it to ensure accurate diff and commit context gathering for the report.

How are code review findings prioritized when preparing a pull request?

Code review findings are prioritized into a severity-ranked report. Deduplicated issues are categorized into Critical items that must be fixed, Warnings, Suggestions, and Positive Notes, helping you focus on blocking vulnerabilities before releasing.

Why does my branch comparison show duplicate issues across security and logic checks?

Branch comparison may surface duplicate issues because parallel reviewers analyze security, correctness, and performance simultaneously. The review process explicitly deduplicates these overlapping findings to produce a consolidated, severity-ranked report.