What problem does it solve? AI agents and developers often claim work is complete, tests pass, or bugs are fixed without actually running the verification commands, leading to broken builds, shipped defects, and lost trust. This Skill eliminates unverified success claims by requiring fresh command output as evidence before any assertion. ## Core Features & Use Cases - Evidence-First Gate: Blocks any completion claim until the relevant verification command (tests, linter, build) has been run and its output read in the current session. - Failure Pattern Detection: Flags rationalizations like "should pass", "linter passed so build works", or trusting agent success reports without checking the VCS diff. - Regression Test Discipline: Enforces the red-green cycle for regression tests, requiring the test to fail when the fix is reverted before it counts as valid. - Use Case: Before committing a bug fix, run the test suite, confirm zero failures in the output, verify the original symptom is resolved, and only then state the fix is complete with the evidence attached. ## Quick Start Before claiming any task is done, run the full verification command, read the complete output including exit code, and only then state the result with that evidence.
