Can I customize the quality thresholds for the audit?

Yes, users can override default settings for parallel worktree limits, severity thresholds, and module lines-of-code (LOC) limits via environment variables or prompts.

What phases are included in the Quality Audit Workflow?

The workflow follows a 6-phase process: project familiarization, parallel auditing, issue assembly, PR generation, PM review, and master reporting.

How does the skill handle large codebases?

It divides the codebase into logical sections and utilizes multiple parallel agents to analyze different modules simultaneously, ensuring high throughput and efficiency.

Does this skill provide automated fixes or just reporting?

It provides both. The workflow identifies issues, creates GitHub tracking issues, and then generates automated Pull Requests in separate worktrees to remediate the findings.

Quality Audit Workflow

Name: Quality Audit Workflow
Author: rysweet

byrysweet

•

Security & Testing

Orchestrates comprehensive, parallel codebase audits with automated GitHub issue creation and PR-based remediation.

The Quality Audit Workflow is a high-performance orchestration skill for Claude Code designed to automate the end-to-end codebase health check process. By deploying specialized parallel agents, it systematically analyzes project architecture, identifies technical debt, and enforces strict coding standards such as module size limits and single responsibility principles. The skill goes beyond simple analysis by automatically generating GitHub issues and corresponding Pull Requests for every finding, followed by a PM-architect review phase to prioritize fixes into a cohesive, actionable master report.

Key Features

01Automated Remediation: Automatically generates GitHub issues and individual fix PRs for every detected problem.

02Consolidated Master Reporting: Creates a central GitHub issue linking all findings and PRs into a prioritized action plan.

0317 GitHub stars

04Parallel Agent Orchestration: Deploys multiple specialized agents for security, optimization, and review across codebase divisions.

05Philosophy Enforcement: Ruthlessly applies standards like the <300 LOC module limit and zero-BS coding principles.

06PM-Architect Integration: Prioritizes and groups PRs by category and severity to ensure a logical fix deployment.

Use Cases

01Technical debt analysis and automated refactoring of legacy codebases.

02Architecture reviews and module quality assessments for complex projects.

03Rapid security and health checks during the onboarding of new repositories.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add rysweet/amplihack quality-audit-workflow

For use in Claude.ai and ChatGPT

Download Skill