What problem does it solve? AI-generated code often claims completion without proof. This Skill acts as the quality gate for the Spec-Driven Development verification phase, proving that an implementation actually matches its specs, design, and task list through real test execution rather than static inspection alone. ## Core Features & Use Cases - Runtime Verification: Executes test, build, and coverage commands and maps every spec scenario to a passing covering test, producing a compliance matrix with COMPLIANT, FAILING, UNTESTED, or PARTIAL statuses. - Strict TDD Auditing: When Strict TDD mode is active, validates the RED-GREEN-REFACTOR evidence from the apply phase, audits assertion quality (tautologies, ghost loops, smoke tests), classifies test layers, and measures changed-file coverage. - Graceful Degradation: Adapts verification depth to available artifacts — tasks only, tasks plus specs, or full proposal/specs/design/tasks — and records skipped dimensions explicitly. - Use Case: After an SDD apply phase finishes a change, the orchestrator delegates to this Skill, which runs the test suite, cross-checks spec scenarios against results, and returns a verdict of PASS, PASS WITH WARNINGS, or FAIL with issues grouped by severity. ## Quick Start Ask the orchestrator to run the SDD verification phase for the current change so it delegates to the sdd-verify sub-agent and returns a verification report.
