review-codebases

Review entire codebases and publish confirmed findings as GitHub issues.

4|2|Updated Feb 25, 2026
One-click install
npx skills add https://github.com/LaiTszKin/apollo-toolkit --skill review-codebases
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-codebases
Source: https://github.com/LaiTszKin/apollo-toolkit/tree/main/review-codebases
Command: npx skills add https://github.com/LaiTszKin/apollo-toolkit --skill review-codebases

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Repository-wide code review workflows that require reading the full codebase before judging, prioritize architecture findings over implementation details, and surface a reproducible set of confirmed findings ready for publication via open-github-issue.

Core Features & Use Cases

  • Enforces a full-repo review before making judgments, citing concrete files and evidence.
  • Prioritizes architecture and system design, then code quality, then edge cases, stopping when a higher tier has confirmed findings.
  • Publishes one GitHub issue per confirmed finding using the open-github-issue dependency, with structured fields and reproducible outputs.
  • Provides a clear output outline: coverage, review tier reached, confirmed findings, publication status, and deferred follow-ups.

Quick Start

Use the review-codebases skill to perform a full repository audit and publish issues for each confirmed finding.

Frequently Asked Questions about review-codebases

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

FAQPage Schema
How do I automate a full repository codebase audit and publish findings to GitHub issues?

Automate a full repository codebase audit by reading the entire project, prioritizing architecture findings over implementation details, and publishing one GitHub issue per confirmed finding with structured, reproducible fields.

What is the best way to perform an architecture review for a large codebase?

An architecture review for a large codebase should read the entire repository first, prioritize system design findings, then code quality, stopping when a higher tier has confirmed findings to ensure governance-ready evaluation.

How does automated code review handle maintainability assessments across an entire project?

Automated maintainability assessment evaluates the entire repository by extracting architecture findings, evaluating code quality, and generating deterministic, reproducible reports suitable for project-level governance.

Can I use this to review large repositories and automatically open GitHub issues for findings?

Yes, you can review large repositories by enforcing a full-repo read before judgment, then using the open-github-issue dependency to publish one structured issue per confirmed architectural or maintainability finding.

Does the code review process prioritize architecture over implementation details?

Yes, the code review process prioritizes architecture and system design first, then code quality, then edge cases, stopping evaluation when a higher tier yields confirmed findings to focus on structural integrity.

What limitations should I expect when automating codebase audits for governance reporting?

Limitations include stopping the review when a higher tier confirms findings, deferring lower-tier edge cases as follow-ups, and restricting publication solely to the open-github-issue format for confirmed findings.