What is a truth score in this skill?

A truth score is a real-time reliability metric on a scale of 0.0 to 1.0 used to assess the quality, security, and correctness of code or agent-performed tasks.

Can I integrate this skill into GitHub Actions?

Yes, the skill provides structured JSON output and CLI commands specifically designed for integration into CI/CD pipelines like GitHub Actions and GitLab CI.

How does the automatic rollback feature work?

If a verification check falls below your defined threshold (default is 0.95), the system automatically uses Git to revert the changes to the last known 'good' state instantly.

What specific criteria does the verification system check?

It evaluates code correctness (syntax/logic), security vulnerabilities, adherence to best practices (SOLID/modularity), performance impact, and documentation completeness.

Verification & Quality Assurance

Name: Verification & Quality Assurance
Author: ricable

byricable

•

安全与测试

Ensures codebase reliability through automated truth scoring, quality verification, and instant git-based rollback systems.

The Verification & Quality Assurance skill provides a robust framework for maintaining high-quality agent outputs and codebase integrity within the Claude Code environment. By implementing a sophisticated truth scoring system (0.0-1.0 scale), automated security and logic checks, and a fail-safe automatic rollback mechanism, it ensures that only production-grade code persists. This skill is essential for developers who need to enforce strict reliability thresholds, monitor agent performance trends, and automate quality gates within their CI/CD pipelines.

主要功能

012 GitHub stars

02Real-time Truth Scoring (0.0-1.0 reliability scale)

03Interactive Web Dashboards with live quality metrics

04Automated Security, Logic, and Best Practice Checks

05Instant Git-based Rollback for failed verifications

06Comprehensive CI/CD Integration for automated quality gates

使用场景

01Automating regressions checks and security audits in high-stakes environments

02Validating AI-generated code against a 0.95 accuracy threshold before merging

03Monitoring long-term performance trends of multiple AI agents across a codebase

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

Download Skill