requesting-code-review

Review git commit range diffs against plan requirements and categorize issues by severity.

Updated Jan 25, 2026
One-click install
npx skills add https://github.com/MINT-IA/MINT --skill requesting-code-review-mint-ia
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: requesting-code-review
Source: https://github.com/MINT-IA/MINT/tree/main/.claude/skills/requesting-code-review
Command: npx skills add https://github.com/MINT-IA/MINT --skill requesting-code-review-mint-ia

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a repeatable process to request precise, actionable code review feedback on a commit range or feature work so issues are caught early and implementations stay aligned with requirements.

Core Features & Use Cases

  • Structured Review Dispatch: Prepare and send a concise review request including BASE_SHA and HEAD_SHA plus an implementation summary to a dedicated reviewer agent.
  • Checklist-Driven Evaluation: Compare changes against plan requirements, assess code quality, architecture, testing, and categorize findings by severity with actionable fixes.
  • Use Cases: Use after each task in subagent-driven development, before merging a major feature, or when stuck to get an independent assessment and prevent regressions.

Quick Start

Dispatch the code-reviewer subagent with BASE_SHA, HEAD_SHA, a short description of what was implemented, and the plan reference to receive a categorized, actionable review.

Frequently Asked Questions about requesting-code-review

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

FAQPage Schema
How do I automate code review for a specific git commit range?

To perform a git commit range code review, provide a reviewer subagent with BASE_SHA and HEAD_SHA references. The subagent evaluates code quality, architecture, and tests against plan requirements, returning categorized issues with file:line references and concrete fixes.

What is checklist-driven code review and when do I need it?

Checklist-driven code review evaluates git diffs against plan requirements to assess code quality, architecture, and testing. You need this process before merging major feature branches, after completing subagent tasks, or when stuck to catch regressions early and maintain implementation alignment.

Can I use an automated code reviewer for pull requests in subagent-driven development?

Yes, you can request a targeted pull request code review in subagent-driven development by dispatching a reviewer subagent. Send the BASE_SHA, HEAD_SHA, implementation summary, and plan reference to receive an independent, categorized assessment with actionable fixes before merging.

What's the best way to ensure code quality before merging a major feature branch?

The best way to ensure code quality before merging a major feature branch is dispatching a structured review request. Provide the BASE_SHA, HEAD_SHA, and implementation summary to a reviewer subagent to compare changes against plan requirements and categorize issues by severity with actionable fixes.

Do I need to provide a plan reference to get actionable code review feedback?

Yes, providing a plan reference is required to get actionable code review feedback. The reviewer subagent uses the stated plan requirements to compare against your BASE_SHA..HEAD_SHA diffs, ensuring the evaluation checks for adherence to the original implementation plan and categorizes issues accurately.

Why does my code review feedback lack specific file and line references?

Code review feedback lacks specific file and line references when the reviewer subagent is not dispatched with the correct BASE_SHA and HEAD_SHA commit range. Providing the exact git diff range ensures the evaluation categorizes issues by severity with precise file:line references for production readiness.