full-repo-review

Review all repository source files and generate a prioritized severity report.

415|44|Updated Mar 18, 2026
One-click install
npx skills add https://github.com/notque/vexjoy-agent --skill full-repo-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: full-repo-review
Source: https://github.com/notque/vexjoy-agent/tree/main/skills/research/full-repo-review
Command: npx skills add https://github.com/notque/vexjoy-agent --skill full-repo-review

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Full-repo reviews are difficult to run thoroughly and consistently, causing teams to miss systemic quality issues or produce shallow, non-prioritized findings.

Core Features & Use Cases

  • Comprehensive 3-wave coverage: Reviews ALL repository source files (optionally scoped to a single directory) rather than only changed files.
  • Deterministic pre-check + evidence-driven grading: Runs a scoring script to flag structural problems before LLM review, then merges results into a single prioritized report.
  • Prioritized backlog instead of auto-fixes: Produces a triage-ready full-repo-review-report.md with severity sections and systemic patterns, avoiding risky large automated patch sets.

Use cases include quarterly codebase health checks, post-refactor verification, onboarding to a new repository, and identifying recurring reliability/quality issues across many components.

Quick Start

Ask the agent to run a full codebase health check and produce a prioritized issue backlog report.

Frequently Asked Questions about full-repo-review

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

FAQPage Schema
How do I run a comprehensive full repository codebase health check?

Run a full repository review to scan all source files, apply deterministic scoring, and execute a 3-wave LLM analysis to generate a prioritized issue backlog. This process evaluates systemic quality across skills, agents, scripts, hooks, and docs.

What is the best way to review an entire codebase for onboarding or post-refactor verification?

Use a full repository review to evaluate every source file and rank repo issues by severity. It merges deterministic pre-checks with evidence-driven LLM grading into a single triage-ready report, ensuring comprehensive coverage for onboarding or post-refactor verification.

How does deterministic scoring work in a full repository quality review?

Deterministic scoring in a full repository review runs a scoring script to flag structural problems before the LLM review begins. These pre-check results are then merged with evidence-driven LLM analysis to create a single prioritized backlog of severity findings.

Can I scope a full repository review to a single directory instead of scanning everything?

Yes, a full repository review can optionally be scoped to a single directory while still scanning all source files within that scope. This allows targeted codebase health checks without reviewing the entire repository structure.

Does a full repository review automatically fix the codebase issues it finds?

No, a full repository review produces a triage-ready report with severity sections and systemic patterns instead of auto-fixes. This avoids risky large automated patch sets and provides a prioritized backlog for manual remediation.

Why should I scan all source files instead of just using git diff for a codebase health check?

Scanning all source files detects systemic quality issues across the entire repository that git diff would miss. A full repository review requires evaluating every file to identify recurring reliability problems across multiple components and generate comprehensive findings.