code-review

Diff branch changes against a base and generate scored findings reports.

226|55|Updated Mar 19, 2026
One-click install
npx skills add https://github.com/Miosa-osa/canopy --skill code-review-miosa-osa
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/Miosa-osa/canopy/tree/main/library/skills/development/code-review
Command: npx skills add https://github.com/Miosa-osa/canopy --skill code-review-miosa-osa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Branch code reviews are slow and error-prone; this skill provides a structured review workflow that scores changes against project guidelines, surfaces categorized findings with severities, and generates LLM-based fix prompts to guide remediation.

Core Features & Use Cases

  • Scored reports with categorized findings and severity levels
  • Auto-generated prompts for fixes that can be executed by an LLM
  • Quality gate: post review to PR if score meets threshold
  • Supports diff-based review between base and target branches, with configurable base and branch
  • Flexible output formats (markdown or JSON) for CI integration

Quick Start

Run the /code-review command to review your current branch against the main baseline and produce a structured report.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I automate code reviews on a pull request branch?

Automate code reviews by diffing changes against a base branch and producing a scored findings report with actionable fix prompts. The review prioritizes logic and tests, generating a final pass/fail verdict for your pull request.

How does a quality gate work for pull request code reviews?

A quality gate evaluates the review score against a threshold and posts the review to the pull request if it meets requirements. This gates merges by surfacing categorized findings and severities directly on the pull request.

Can I generate code review reports in JSON format for CI workflows?

Yes, code review reports support flexible output formats including JSON and markdown for CI integration. The report contains categorized findings, severity levels, and auto-generated LLM fix prompts to guide remediation.

What is the best way to get actionable fix prompts for code review findings?

Generate actionable fix prompts by running a structured code review that scores changes against project guidelines. The review automatically categorizes findings by severity and produces LLM-based prompts to guide remediation.

Does automated code review work across multiple programming languages?

Yes, automated code review applies across languages by diffing branch changes against a configurable base. It prioritizes logic and tests to produce a scored report with a pass/fail verdict regardless of language.

Why use scored findings reports instead of manual code reviews?

Scored findings reports replace slow, error-prone manual reviews by structuring the review workflow against project guidelines. This surfaces categorized findings with severities and generates actionable fix prompts for faster remediation.