Can I use this in my GitHub Actions?

Yes, the skill supports structured JSON output and CLI commands specifically designed for easy integration into GitHub Actions, GitLab CI, and other automation pipelines.

What is a Truth Score?

A Truth Score is a metric from 0.0 to 1.0 that evaluates the reliability, correctness, and security of code or agent tasks, where 1.0 represents production-ready excellence.

What coding standards does it check?

It evaluates code correctness, TypeScript types, SOLID principles, security vulnerabilities, performance complexity, and documentation completeness.

Is the verification threshold customizable?

Yes, you can adjust the threshold via CLI or configuration, allowing you to set strict requirements (0.99) for production or more lenient ones (0.90) for development.

How does the automatic rollback work?

If a code change fails a verification check or falls below the specified accuracy threshold (default 0.95), the skill uses Git to instantly revert the codebase to its last known good state.

Verification & Quality Assurance

Name: Verification & Quality Assurance
Author: ricable

byricable

•

Security & Testing

Ensures codebase reliability through automated truth scoring, security audits, and an instant rollback system for failed quality checks.

The Verification & Quality Assurance skill is a specialized tool for developers using Claude Flow who need to maintain high standards for AI-generated code. It introduces a sophisticated 'Truth Scoring' system that evaluates agent outputs on a scale of 0.0 to 1.0, covering metrics like logic flow, security vulnerabilities, and adherence to best practices. By integrating directly with Git, the skill provides a safety net with automatic rollbacks for any changes falling below a 0.95 accuracy threshold, making it ideal for maintaining production-grade reliability in agentic workflows and CI/CD pipelines.

Key Features

01Automated security, logic, and best practice verification

02Deep CI/CD integration with JSON and HTML reporting

032 GitHub stars

04Interactive web dashboard with live quality trends

05Instant Git-based rollback for failed code changes

06Real-time Truth Scoring (0.0-1.0) for reliability metrics

Use Cases

01Automating security and style audits in continuous integration pipelines

02Monitoring agent performance trends across long-running development tasks

03Ensuring AI-generated code meets production standards before merging

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ricable/ultimate-ai-agent verification-quality

For use in Claude.ai and ChatGPT

Key Features

01Automated security, logic, and best practice verification

02Deep CI/CD integration with JSON and HTML reporting

032 GitHub stars

04Interactive web dashboard with live quality trends

05Instant Git-based rollback for failed code changes

06Real-time Truth Scoring (0.0-1.0) for reliability metrics

Use Cases

01Automating security and style audits in continuous integration pipelines

02Monitoring agent performance trends across long-running development tasks

03Ensuring AI-generated code meets production standards before merging

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ricable/ultimate-ai-agent verification-quality

For use in Claude.ai and ChatGPT