review

Review staged Git diffs against seven BotAssist code quality criteria.

Updated Apr 7, 2026
One-click install
npx skills add https://github.com/amotion-ai/botassist --skill review-amotion-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/amotion-ai/botassist/tree/main/.claude/skills/review
Command: npx skills add https://github.com/amotion-ai/botassist --skill review-amotion-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It provides an automated, read‑only analysis of code changes in BotAssist projects, enforcing security, quality, and consistency rules before a commit is allowed.

Core Features & Use Cases

  • Seven‑criteria review covering correctness, security, blast radius, tests, conventions, error handling, and frontend/backend consistency.
  • Git integration works on staged diffs or specific file paths, automatically gathering related files based on the project dependency map.
  • Actionable report with severity classifications (MUST FIX, SHOULD FIX, NITPICK) and a clear verdict (APPROVE, APPROVE WITH SUGGESTIONS, REQUEST CHANGES) to guide reviewers.

Quick Start

Invoke the /review command with optional arguments to analyze your code changes according to BotAssist’s quality criteria.

Frequently Asked Questions about review

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

FAQPage Schema
How do I review staged Git diffs for security and quality issues?

To review staged Git diffs for security and quality, this Skill analyzes changes against seven criteria including correctness, security, blast radius, tests, conventions, error handling, and frontend/backend consistency, producing a severity-rated report with approval verdicts.

What criteria are evaluated during an automated code review?

Automated code review evaluates seven criteria: correctness, security, blast radius, tests, conventions, error handling, and frontend/backend consistency. The analysis generates actionable severity classifications like MUST FIX, SHOULD FIX, and NITPICK to guide commit approvals.

Can I scan specific file paths instead of staged Git changes?

Yes, you can scan specific file paths instead of staged Git changes. The Skill operates on either staged Git diffs or specified files, automatically gathering related files based on the project dependency map to ensure comprehensive analysis.

Does this code review tool enforce project-specific conventions?

Yes, this code review tool enforces project-specific conventions by reading the CLAUDE.md file for rules. It scans related files and evaluates code changes for compliance with these defined standards before allowing a commit.

What severity classifications are used in the code review report?

The code review report uses three severity classifications: MUST FIX, SHOULD FIX, and NITPICK. These classifications accompany a clear verdict of APPROVE, APPROVE WITH SUGGESTIONS, or REQUEST CHANGES to guide reviewers.

When do I need an automated read-only analysis of code changes?

You need an automated read-only analysis of code changes when enforcing security, quality, and consistency rules before a commit is allowed. This ensures BotAssist projects maintain compliance standards without manually inspecting every file.