What is the default truth score threshold for verification?

The skill uses a default threshold of 0.95. Any code or agent output scoring below this level can be configured to trigger an automatic rollback or a failure alert.

Can I integrate these metrics into GitHub Actions?

Yes, the skill is designed for CI/CD. It can output structured JSON, allowing you to fail builds if the truth score drops below your required threshold.

What aspects of the code are evaluated during a check?

The system evaluates code correctness (syntax/logic), security (vulnerabilities/secrets), best practices (SOLID/modularity), performance (complexity), and documentation quality.

Does this skill require Git to function?

While truth scoring and verification checks work independently, the automatic rollback feature requires a Git repository to instantly revert changes to the last known good state.

Verification & Quality Assurance

Name: Verification & Quality Assurance
Author: Activer007

byActiver007

0•

安全与测试

Ensures codebase reliability and agent output quality through automated truth scoring, security verification, and git-powered rollbacks.

This skill acts as a robust safety layer for AI-driven development by providing a sophisticated verification system for Claude-powered agents. It evaluates outputs against a high-accuracy 0.95 threshold, performing deep analysis of code correctness, security vulnerabilities, and adherence to best practices. With built-in automatic rollback capabilities, real-time monitoring dashboards, and seamless CI/CD integration, it empowers developers to maintain production-grade standards while automating complex tasks and swarm operations.

主要功能

01Real-time Truth Scoring (0.0-1.0 scale) for code and agent reliability

02Automated verification of security, performance, and SOLID principles

03Comprehensive CI/CD integration with JSON, HTML, and CSV reporting

04Interactive web-based dashboards for live quality metrics and trends

05Instant Git-based rollback for any changes failing quality thresholds

060 GitHub stars

使用场景

01Monitoring long-term agent performance and codebase health trends

02Establishing quality guardrails for autonomous AI agent workflows

03Implementing automated security and correctness checks in CI/CD pipelines

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add activer007/ordinary-claude-skills verification-quality-assurance

For use in Claude.ai and ChatGPT

Download Skill