audit

Audits codebases through twelve concern lenses covering security, tech debt, design, data models, and scale.

Updated May 3, 2026
One-click install
npx skills add https://github.com/Claudfather/clauDNA --skill audit-claudfather
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: audit
Source: https://github.com/Claudfather/clauDNA/tree/main/skills/audit
Command: npx skills add https://github.com/Claudfather/clauDNA --skill audit-claudfather

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Codebases accumulate hidden risks — security vulnerabilities, technical debt, stale documentation, design inconsistencies, data model mismatches, and scaling bottlenecks — that are hard to detect without a systematic, evidence-driven review process. ## Core Features & Use Cases - Twelve audit lenses: security, tech-debt, docs, design, access-path, data-model, data-model-redesign, frontend-perf, repo-health, system, cache, and scale — each with its own depth procedure and severity classification. - Flexible output routing: present findings in chat, file them as deduplicated GitHub issues via the publish pipeline, or generate phased remediation plan documents. - Autonomous mode: run supported lenses non-interactively with --auto for CI and headless environments, emitting structured JSON results. - Use Case: Run /claudna:audit security --auto --output github in CI to scan for injection flaws and secrets exposure, automatically filing prioritized GitHub issues for each confirmed finding. ## Quick Start Ask Claude to run an audit on your repository, for example by saying: audit this codebase for security vulnerabilities and file the findings as GitHub issues.

Frequently Asked Questions about audit

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

FAQPage Schema
How do I audit a codebase for security vulnerabilities?

Run the audit skill with the security lens, which scans for injection flaws, auth and authorization gaps, secrets exposure, and unsafe dependencies. Findings are severity-ranked with file and line evidence, and can be filed as GitHub issues with --output github.

What audit types does this skill support?

It supports twelve lenses: security, tech-debt, docs, design, access-path, data-model, data-model-redesign, frontend-perf, repo-health, system, cache, and scale. Each lens has its own procedure, severity model, and output targets.

Can I run a codebase audit non-interactively in CI?

Yes, lenses marked auto-capable (security, tech-debt, docs, access-path, frontend-perf, cache, scale) run with --auto, skip user gates, and emit a structured JSON result. Interactive-only lenses like design and data-model-redesign return a blocked outcome instead.

What is the difference between the data-model and data-model-redesign lenses?

The data-model lens is a fast fit-audit of the current schema against the code that uses it. The data-model-redesign lens evaluates whether the model should be rebuilt, compares at least three candidate architectures including incremental repair, and plans a staged migration.

Does the audit skill modify my code?

No, the audit lenses produce findings, plans, and issues rather than code changes. Remediation plans are handed off to the separate build skill, and the docs lens only edits documentation files, never source code.