skills.rest
anthril
anthrilOFFICIAL
Explore all skills by this author →
WebsiteGitHub

application-audit

Orchestrate a multi-agent audit of Next.js, React, and Supabase codebases.

3|Updated Apr 1, 2026
Search Tags:#markdown#validation#audit#mcp#reporting#mult-agent#software-audit
One-click install
npx skills add https://github.com/anthril/official-claude-plugins --skill application-audit
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: application-audit
Source: https://github.com/anthril/official-claude-plugins/tree/main/plugins/software-development/skills/application-audit
Command: npx skills add https://github.com/anthril/official-claude-plugins --skill application-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Orchestrates a multi-agent audit of a codebase and produces a validated, evidence-backed report.

Core Features & Use Cases

  • Orchestrates nine domain-specific auditors (frontend, backend, security, leak-detection, etc.) against a codebase to surface findings with evidence.
  • Validates findings via a centralized validator and emits a final, ranked REPORT.md plus JSON sidecar.
  • Supports an open-question gate to capture unknowns and maintain traceability, while keeping source read-only.

Quick Start

Point the skill at a project directory and run the audit to produce a final REPORT.md along with per-agent reports under .anthril/audits/<ID>/.

Frequently Asked Questions about application-audit

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

FAQPage Schema
How do I run an evidence-backed codebase audit for a Next.js and Supabase project?▼

You can audit a Next.js, React, and Supabase codebase by pointing the skill at your project directory, which orchestrates nine domain-specific auditors to produce a validated REPORT.md and JSON sidecars.

What is a multi-agent software audit and how does it validate findings?▼

A multi-agent software audit deploys nine specialized auditors across a codebase to surface findings, which a centralized validator then checks to ensure accuracy and maintain traceability before emitting a final report.

Can I audit my codebase without modifying the source files?▼

Yes, the audit enforces a strictly read-only source policy, allowing the multi-agent system to analyze code and generate reports under .anthril/audits/ without altering your original project files.

Does the codebase audit support projects outside of Next.js and Supabase?▼

The audit is specifically designed to apply to Next.js, React, and Supabase applications, orchestrating domain-specific checks tailored to that stack to ensure accurate and validated reporting.

What is the best way to capture unknowns and maintain traceability during a software audit?▼

The best way to capture unknowns during a software audit is using an open-question gate mechanism, which records unresolved questions alongside evidence-backed findings to maintain full traceability in the final report.

What output formats are generated when auditing a codebase for security and leak detection?▼

Auditing a codebase generates a final ranked REPORT.md file alongside JSON sidecars, containing validated findings from specialized agents including security and leak-detection auditors.

Related Skills

Explore More Software Engineering Skills→
XWIlluDeluXWIlluDelu

project-setup

Link shared skills and fragments into a repository and generate AGENTS.md with alias symlinks.

Community
Intermediate
Logisoft-webLogisoft-web

vulnerability-scanner

Scan code repositories for vulnerabilities in dependencies, secrets, and configurations.

Community
Advanced
ryasrkryasrk

memory-blackboard

Implement a shared memory blackboard protocol for multi-agent session communication.

Community
Intermediate
OneWave-AIOneWave-AI
245

code-review-pro

Review code for security vulnerabilities, performance bottlenecks, and maintainability issues.

Official
Basic