概要
The Verification Loop skill provides a comprehensive six-phase quality assurance framework for Claude Code sessions, enabling AI agents to self-validate work before submission. By systematically executing build processes, language-specific type-checking (TypeScript/Python), linting, and security scans, it ensures that all code changes meet professional standards. The skill includes a rigorous test suite evaluator with coverage tracking and a Git diff review phase, culminating in a standardized report that confirms whether a feature is truly ready for a pull request.