Agent skill
audit
Run comprehensive codebase audit for gaps, deprecated code, TODOs, FIXMEs, architectural anti-patterns, type issues, and code smells. Use when user asks to audit code, find issues, check code quality, or identify architectural problems.
Install this agent skill to your Project
npx add-skill https://github.com/Cygnusfear/claude-stuff/tree/main/skills/audit
SKILL.md
Codebase Audit
Instructions
Perform a comprehensive, systematic audit of the codebase to identify quality issues, architectural problems, and technical debt.
Phase 1: Discovery & Planning
- Identify scope - Determine which files/directories to audit based on user request
- Create comprehensive file list - Use Glob to find all relevant files
- Initialize todo list - Create a todo with one item per file to audit
- Set up audit report - Create structured markdown report at
.audit/audit-report-[timestamp].md
Phase 2: Automated Analysis
Run automated tools to supplement manual review:
- TypeScript compiler diagnostics
- ESLint (if configured)
- Grep for common patterns: TODO, FIXME, HACK, XXX, @deprecated
Phase 3: Systematic File Review
For EACH file in the todo list:
-
Read and analyze the file thoroughly
-
Check for issues in these categories:
- Deprecations: Deprecated APIs, patterns, or code marked for removal
- TODOs/FIXMEs: Unfinished work or known issues
- Architectural anti-patterns:
- God objects/classes
- Circular dependencies
- Tight coupling
- Violation of SOLID principles
- Inconsistent patterns
- Type issues:
- Use of
anyorunknown - Missing type annotations
- Incorrect type usage
- Type casts that hide issues
- Use of
- Code smells:
- Duplicated code
- Long functions/classes
- Complex conditionals
- Dead code
- Magic numbers/strings
- Poor naming
-
Assign severity to each finding:
- CRITICAL: Breaks functionality, security issues, data corruption risks
- HIGH: Architectural violations, major maintainability issues
- MEDIUM: Code smells, minor anti-patterns, missing types
- LOW: Style issues, minor TODOs, cosmetic improvements
-
Check for cross-file patterns - As you review, note patterns that appear across multiple files
-
Update report - Add findings to the structured report
-
Mark file as completed in todo list
Phase 4: Cross-File Analysis
After reviewing all individual files:
- Identify systemic patterns - Issues that appear across multiple files
- Architectural assessment - Overall system architecture health
- Dependency analysis - Check for circular dependencies or coupling issues
- Consistency check - Verify naming conventions, patterns are followed
Phase 5: Validation & Summary
-
Run final checks:
- TypeScript type check (
tsc --noEmitor similar) - Linting (
npm run lintor similar) - Build process if applicable
- TypeScript type check (
-
Generate executive summary:
- Total issues by category
- Total issues by severity
- Top 10 most critical findings
Audit Report Structure
# Audit Report - [Date]
## Executive Summary
- **Files Audited**: X
- **Total Issues Found**: Y
- **Critical**: A | **High**: B | **Medium**: C | **Low**: D
## Top 10 Critical Findings
1. [Issue description] - Severity: CRITICAL - File: path/to/file.ts:line
## Issues by Category
### Deprecations
- [Issue] - Severity - File:line
### TODOs/FIXMEs
- [Issue] - Severity - File:line
### Architectural Anti-Patterns
- [Issue] - Severity - File:line
### Type Issues
- [Issue] - Severity - File:line
### Code Smells
- [Issue] - Severity - File:line
## Cross-File Patterns
- [Pattern description and affected files]
## Automated Tool Results
- TypeScript diagnostics summary
- ESLint results summary
Critical Principles
- NEVER skip files - Audit every file in the todo list
- NEVER edit files during audit - This is read-only analysis
- NEVER provide recommendations - Only identify and report problems
- NEVER create action plans - That's a separate responsibility
- DO use memory/pinboard - Store context as you discover patterns
- DO be thorough - Think critically about each file
- DO be objective - Report what you find, not what to do about it
- DO track progress - Keep todo list updated in real-time
- DO find all relevant files - If you discover new files that should be audited, add them to the todo
Dynamic File Discovery
If during audit you discover additional files that should be reviewed:
- Add them to the todo list immediately
- Continue systematic review
- Ensure no stone is left unturned
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
brainstorming
You MUST use this, UNLESS the human says otherwise, before any EXTENDED creative work (not simple execution) - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
axiom-audit
ONLY USE WHEN REPO IS USING AXIOM LOGGING. Audit Axiom logs to identify and prioritize errors and warnings, research probable causes, and flag log smells. Use when user asks to check Axiom logs, analyze production errors, investigate log issues, or audit logging patterns.
test-driven-development
Use when implementing any feature or bugfix, before writing implementation code
create-plan
Create comprehensive implementation plan in .plans directory based on analysis or report. Use when user asks to create a plan, plan implementation, design a solution, or structure work for a feature/refactor/fix.
video-explorer
This skill should be used when analyzing video files. Claude cannot process video directly, so this skill extracts frames hierarchically - starting with a quick overview, then zooming into regions of interest with higher resolution and temporal density. Use when asked to watch, analyze, review, or understand video content.
using-superpowers
Guidelines for using skills effectively - load relevant skills before complex tasks, not every message
Didn't find tool you were looking for?