What is Truth Scoring in Claude Code?

Truth Scoring is a reliability metric on a 0.0 to 1.0 scale that evaluates code or agent tasks based on syntax validation, logic flow, security standards, and adherence to best practices.

Does it provide visual reports for team reviews?

Yes, it includes a built-in interactive dashboard and can export HTML reports with charts showing quality trends, agent performance, and verification history.

Is the quality threshold customizable?

Absolutely. You can adjust the threshold globally or per environment, such as requiring 0.99 for production-critical code and 0.90 for development experiments.

How does the automatic rollback feature work?

If a code change fails to meet the defined quality threshold (default 0.95), the skill uses Git integration to instantly revert the codebase to its last known good state, preventing broken code from persisting.

Can I integrate this skill with my existing CI/CD pipeline?

Yes, the skill provides structured JSON and CSV outputs specifically designed for integration with GitHub Actions, GitLab CI, and other automation platforms to enforce quality gates.

Claude Code Verification & QA Skill | High-Quality AI Coding

The Verification & Quality Assurance skill is a specialized tool for developers using Claude Code who require production-grade reliability from AI-generated code. It implements a sophisticated 'Truth Scoring' system that evaluates code correctness, security, and best practices on a 0.0-1.0 scale. By enforcing a customizable accuracy threshold and providing automatic Git-based rollbacks for failed checks, this skill prevents regressions and maintains codebase integrity. It seamlessly integrates into CI/CD pipelines and provides a real-time dashboard for monitoring agent performance and quality trends throughout the software development lifecycle.

주요 기능

01Real-time Truth Scoring (0.0-1.0) for agent tasks and code quality

02Interactive web-based dashboard for live monitoring and trend analysis

03Seamless CI/CD integration for GitHub Actions and GitLab CI pipelines

04Automated verification checks for security, performance, and best practices

05206 GitHub stars

06Intelligent automatic rollback system to revert changes below quality thresholds

사용 사례

01Automating security and performance audits within the agentic development workflow

02Monitoring and benchmarking the reliability of different coding agents over time

03Ensuring AI-generated code meets strict production quality standards before merging

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add proffesor-for-testing/agentic-qe verification-quality

For use in Claude.ai and ChatGPT

Download Skill

주요 기능

01Real-time Truth Scoring (0.0-1.0) for agent tasks and code quality

02Interactive web-based dashboard for live monitoring and trend analysis

03Seamless CI/CD integration for GitHub Actions and GitLab CI pipelines

04Automated verification checks for security, performance, and best practices

05206 GitHub stars

06Intelligent automatic rollback system to revert changes below quality thresholds

사용 사례

01Automating security and performance audits within the agentic development workflow

02Monitoring and benchmarking the reliability of different coding agents over time

03Ensuring AI-generated code meets strict production quality standards before merging

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add proffesor-for-testing/agentic-qe verification-quality

For use in Claude.ai and ChatGPT

Download Skill

Verification & Quality Assurance for Claude Code

Verification & Quality Assurance for Claude Code

주요 기능

사용 사례

주요 기능

사용 사례