Verification & Quality Assurance

Automate verification, truth scoring, and rollback in CI/CD pipelines.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/bjpl/hablas --skill verification-quality-assurance-bjpl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Verification & Quality Assurance
Source: https://github.com/bjpl/hablas/tree/main/.claude/skills/verification-quality
Command: npx skills add https://github.com/bjpl/hablas --skill verification-quality-assurance-bjpl

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires claude-flow, git.

What problem does it solve?

Ensuring the quality and reliability of AI-generated code and agent outputs is critical but often manual. This skill provides a comprehensive system for real-time truth scoring, automated quality checks, and instant rollback of changes that fail to meet defined thresholds, guaranteeing high-quality results.

Core Features & Use Cases

  • Truth Scoring: Real-time reliability metrics (0.0-1.0) for code, agents, and tasks.
  • Automated Verification Checks: Validate code correctness, security, best practices, and documentation.
  • Automatic Rollback: Instantly revert changes that fail verification (default threshold: 0.95).
  • Use Case: Integrate this skill into your CI/CD pipeline to automatically verify every code change. If a pull request introduces code that scores below a 0.95 truth score, the system can automatically block the merge and even roll back the changes, preventing low-quality code from entering the main branch.

Quick Start

View current truth scores and then run a verification check on your current working directory. npx claude-flow@alpha truth npx claude-flow@alpha verify check

Frequently Asked Questions about Verification & Quality Assurance

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automatically verify code quality and block low-quality changes in CI/CD?

Verification & Quality Assurance automates code quality checks with truth scoring (0.0-1.0 scale) and blocks merges below a 0.95 threshold. Integrate it into your CI/CD pipeline to validate code correctness, security, and best practices on every pull request, preventing degraded code from reaching main.

What is truth scoring and how does it measure code reliability?

Truth scoring generates real-time reliability metrics on a 0.0-1.0 scale for code, agents, and tasks. The system evaluates code correctness, security compliance, and adherence to best practices, then assigns a single score that determines whether changes pass verification or trigger automatic rollback.

Can I automatically rollback code changes that fail quality checks?

Yes. When code scores below the 0.95 truth threshold, the skill automatically reverts changes to the previous Git state. This instant rollback prevents low-quality code from persisting and keeps your codebase reliable without manual intervention.

Do I need Git and Node.js 18+ to use this verification system?

Yes. This skill requires Git repositories for rollback capability, Node.js 18 or higher, and Claude Flow installation. These prerequisites enable automated verification checks and instant reversion of failed changes across your codebase.

How do I export verification results in different formats?

The skill supports JSON, CSV, and HTML export options for verification reports and truth scores. Use the CLI commands `truth`, `verify`, and `rollback` to generate exports in your preferred format for analysis, compliance, or team review.

Can I monitor AI agent output quality in production with automated checks?

Yes. The skill applies truth scoring and verification checks to AI agent outputs within live CI/CD pipelines. Real-time monitoring ensures agent reliability by catching quality failures early and triggering automatic rollback when scores fall below threshold.