Run a multi-faceted quality and architecture review on your code.

Install

mkdir -p .claude/skills/audit-alo2118 && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/14669" && unzip -o skill.zip -d .claude/skills/audit-alo2118 && rm skill.zip

Installs to .claude/skills/audit-alo2118

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Run a comprehensive audit of a specific area of the codebase using 3 parallel expert agents. Use when asked to audit, review deeply, or check quality of a module/page/feature.
175 chars✓ has a “when” trigger
Advanced

Key capabilities

  • Dispatch parallel agents for bug hunting
  • Dispatch parallel agents for UX review
  • Dispatch parallel agents for architecture review
  • Consolidate findings from multiple agents
  • Prioritize issues by criticality
  • Re-audit code after fixes

How it works

The skill dispatches three specialized agents (bug-hunter, UX-reviewer, arch-reviewer) to analyze a specified codebase area, consolidates their findings, and prioritizes issues for remediation.

Inputs & outputs

You give it
A specific area of the codebase to audit, and optionally, context about recent changes
You get back
A consolidated list of critical, major, and minor findings from bug-hunter, UX-reviewer, and arch-reviewer agents

When to use audit

  • Reviewing module quality
  • Identifying architectural issues
  • Auditing code patterns

About this skill

Run a comprehensive audit of: $ARGUMENTS

Process

1. Dispatch 3 parallel agents (mandatory)

Launch these 3 agents simultaneously using the Agent tool:

  • bug-hunter agent — Analyze the target area for bugs, data loss risks, Supabase query safety, race conditions
  • ux-reviewer agent — Check UX consistency, accessibility, style constant usage, sibling comparison
  • arch-reviewer agent — Check architecture, file ownership, performance, convention compliance

Each agent must receive:

  • The specific area/files to analyze
  • Context about what was recently changed (if applicable)

2. Consolidate findings

Merge results from all 3 agents into a single ranked list:

  • CRITICAL — Data loss, crashes, security issues
  • MAJOR — Incorrect behavior, accessibility violations, broken patterns
  • MINOR — Style inconsistencies, minor improvements

3. Fix all issues, starting from critical

4. Re-audit after fixes (mandatory)

Run the 3 agents again on the fixed code to verify nothing was missed and no regressions introduced.

5. Build check

Run npm run build — must pass with zero errors.

When not to use it

  • When a complete audit of a specific codebase area is not required
  • When the user does not want to use multiple expert agents for review
  • When the task is not related to code quality or bug fixing

Limitations

  • Requires the ability to launch parallel agents.
  • The audit is limited to the areas and contexts provided.
  • Requires `npm run build` to pass with zero errors after fixes.

How it compares

This skill automates a multi-faceted code audit using parallel expert agents, providing a complete review across different quality dimensions, unlike a single-agent or manual code review.

Compared to similar skills

audit side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
audit (this skill)03moNo flagsAdvanced
python-testing-patterns772moReviewIntermediate
dependency-upgrade265moReviewIntermediate
test-cases577moNo flagsBeginner

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry