hive-tdd-pipeline

You can automate the TDD Red-Green cycle by executing a pipeline that generates failing tests, implements code to pass them, and runs verification. This skill enforces agent isolation between test writers and implementers to prevent context pollution.

Agent isolation in TDD pipelines enforces strict separation between test writers and code implementers. This separation prevents context pollution, ensuring the implementation agent does not artificially tailor tests to pass without meeting true specifications.

You add mutation and property-based testing to CI/CD by running a cross-verify phase after implementation. This phase executes mutation tests, runs property-based checks, and conducts cross-model reviews to validate code robustness comprehensively.

Yes, you can enforce test tampering detection during the Green phase. The pipeline monitors the implementation process within a limited iteration count to detect if the AI agent has improperly modified generated tests to force a passing state.

Yes, the pipeline automatically generates both example-based and property-based tests during the Red phase. It requires all generated tests to fail initially, ensuring the subsequent implementation phase adheres strictly to the provided SPEC document.