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, broken trust, and wasted rework. This Skill blocks any completion claim until fresh command output proves it. ## Core Features & Use Cases - The Iron Law gate: Requires identifying, running, and reading the full verification command output before any success claim, commit, PR, or task closure. - Failure pattern tables: Maps common claims (tests pass, linter clean, bug fixed, agent completed) to the exact evidence required and what does not count as proof. - Rationalization and red-flag detection: Lists excuses like "should work now" or "agent said success" and counters each with the required verification action. - Closing Pass checklist: Walks a five-item Definition of Done (debt, knowledge, board, review, evidence) with a verdict per item before declaring completion. - Use Case: Before creating a PR, the agent runs the full test suite, confirms 0 failures in the output, verifies the regression test fails without the fix, checks the ticket status, and only then reports completion with evidence. ## Quick Start Before you claim this task is done, apply the verification-before-completion gate: run the full test command, read the output, and report the result with evidence.
