verification-before-completion

To enforce verification before completion, identify the exact proving command, execute it freshly, check the exit code and full output, then provide explicit evidence of the results when opening the pull request. This prevents false or premature claims of completion.

A verification gate is a five-step process to identify, run, read, verify, and claim completion based on evidence. It ensures developers and reviewers independently verify test runs, linters, builds, and regression checks before reporting success.

To verify code review changes before committing, run the specified test or build command, inspect the exit code and full output, and attach the test output or cite exact pass counts as explicit evidence of success.

Yes, this verification process applies to automated agents and regression checks. It requires identifying the proving command, executing it freshly, checking exit codes and outputs, and providing explicit evidence before claiming work is complete.

Reading full command outputs is necessary because exit codes alone may not reveal hidden failures or warnings. The verification process requires checking both the exit code and the complete output to ensure accurate evidence before asserting success.

The limitation of enforcing a verification gate is that it requires manually identifying the exact proving command for each change scenario. It does not automatically select tests or builds, meaning developers must specify and execute the appropriate verification commands themselves.