audit-runner

Orchestrate multi-agent code audits and generate reports with action items.

Updated Feb 10, 2026
One-click install
npx skills add https://github.com/ojallington/boardclaude --skill audit-runner
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: audit-runner
Source: https://github.com/ojallington/boardclaude/tree/main/skills/audit-runner
Command: npx skills add https://github.com/ojallington/boardclaude --skill audit-runner

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates the process of evaluating codebases by deploying specialized AI agents to assess quality, identify issues, and suggest improvements, streamlining the development and review cycle.

Core Features & Use Cases

  • Multi-Agent Code Audits: Leverages a configurable panel of AI agents (e.g., architecture, product, code quality) to perform comprehensive evaluations.
  • Closed-Loop Improvement: Facilitates an audit-fix-re-audit cycle to track and demonstrate project improvement over time.
  • Configurable Evaluation Panels: Allows users to define custom agent teams and evaluation criteria via YAML configuration.
  • Real-time Validation: Integrates with common development tools (tsc, jest, eslint, prettier) to validate findings.
  • Use Case: A development team can use this Skill to perform a full audit of their new feature branch, identify critical bugs and architectural flaws, automatically generate fixes, and then re-audit to confirm the improvements before merging.

Quick Start

Run a full panel audit on the current codebase.

Frequently Asked Questions about audit-runner

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

FAQPage Schema
How does multi-agent automated code review work?

Automated code audit tools deploy configurable panels of specialized AI agents to evaluate codebases, identify architectural flaws and bugs, and synthesize findings into detailed reports.

How do I configure custom evaluation criteria for AI code audits?

You configure custom AI agent panels and evaluation criteria by defining them within a YAML configuration file, enabling tailored codebase assessments for your specific project requirements.

Can I integrate automated code audits with standard development tools?

Yes, automated code audits integrate with common development tools like tsc, jest, eslint, and prettier to validate findings, test fixes, and confirm improvements before merging.

What is a closed-loop code audit improvement cycle?

A closed-loop code audit improvement cycle is an audit-fix-re-audit process that evaluates code, generates fixes, and re-runs the audit to track and validate project improvements over time.

What is the best way to evaluate a new feature branch before merging?

The best way to evaluate a feature branch before merging is running a full panel AI code audit to identify critical bugs, generate fixes, and validate improvements using integrated development tools like eslint and jest.

What are the limitations of using AI agents for codebase evaluation?

Limitations of AI codebase evaluation include the potential for synthesized agent findings requiring validation via tools like jest and eslint, and the need for manual review of generated action items before implementation.