verification-before-completion

To prevent false completion claims, you must require fresh verification evidence before marking work done. This involves running full validation commands, checking their exit status and failure counts, and reporting only evidence-backed results to prove completion.

Evidence-first validation is the process of running a relevant command and reading its actual output before claiming success. It ensures that testing, builds, and regression checks are confirmed with real command results rather than assumptions.

You verify test results by executing the exact validation commands and analyzing the output. This process enforces checking the exit status and counting any failures, ensuring you only report success when backed by fresh command evidence.

Yes, this verification approach works for pre-commit checks and unit tests. It is designed to confirm linting, builds, and regression tests with actual results, making it useful when preparing a commit, opening a pull request, or handing work off.

The best way to prove a build is finished is to enforce fresh verification evidence before making success claims. You must run full validation commands, check for failure counts, and demonstrate success strictly through actual command output.

You need to run validation commands whenever a task's success must be demonstrated, such as after bug fixes, unit tests, or pull request preparations. It prevents false completion claims by requiring actual command output evidence before marking work done.