What problem does it solve? AI agents and developers often claim work is done, tests pass, or bugs are fixed without actually running the verification command, leading to broken builds, false status reports, and lost trust. This Skill blocks any completion claim until fresh command output proves it. ## Core Features & Use Cases - The Iron Law gate: No completion claims without fresh verification evidence — identify the proving command, run it fully, read the output and exit code, then claim only what the output confirms. - Failure pattern tables: Maps common claims (tests pass, build succeeds, bug fixed, dev server alive) to the exact evidence required versus insufficient proxies like previous runs or linter output. - Rationalization prevention: Counters excuses like "should work now", "just this once", or "agent said success" with mandatory re-verification, including red-green regression test cycles and VCS diff checks for delegated agent work. - Use Case: Before telling a stakeholder that a bug fix is complete, run the failing test, confirm it passes, revert the fix to confirm the test fails, restore it, and only then report the verified result. ## Quick Start Before claiming any task is finished, invoke this skill to identify the command that proves completion, run it, and report the result with the actual output as evidence.
