Enforces strict, evidence-based verification protocols before any task is marked as complete or fixed.
The Task Completion Verifier skill establishes a rigorous 'Iron Law' for software development, preventing premature or unsubstantiated completion claims. It transforms the AI from an agent that assumes success into one that proves it through fresh execution data, logs, and exit codes. By requiring explicit verification steps—such as running full test suites, build commands, and TDD red-green cycles—this skill ensures that every fix, feature, or refactor is backed by tangible evidence rather than assumptions or partial checks.
Características Principales
01Evidence-based completion protocol
020 GitHub stars
03Mandatory fresh verification execution
04Multi-claim validation for tests, builds, and linters
05TDD Red-Green cycle enforcement
06Rationalization and 'Red Flag' detection
Casos de Uso
01Verifying bug fixes before submitting a pull request
02Ensuring regression tests fail before they pass in TDD workflows
03Validating build success after complex dependency changes
