quality-validation

To enforce evidence before claims of code completion, the skill mandates running specific verification commands and presenting their output. This ensures task success is backed by fresh execution results rather than assumptions.

Mandatory verification for task completion requires independently executing commands to prove a bug is fixed. It prevents false positives by demanding fresh test evidence before any success status is declared.

To prevent false positives when marking agent tasks as complete, you must enforce a strict validation policy. This requires independently executing verification commands to generate fresh output before claiming any status.

Task verification requires running the test suite again to distinguish between actual successful validation and mere assumptions based on past results. This process enforcement maintains trust by proving zero failures.

Code review validation does not work without independent command execution. The process requires generating fresh output from specific commands to clearly distinguish between successful verification and unverified assumptions.

The limitation of relying on past results for quality assurance is the risk of premature task completion claims. Without fresh execution and validation of commands, you cannot prevent false positives or maintain trust.