verification-before-completion

Verification before completion requires running your full test suite, build command, and linter locally, then confirming the output shows all checks pass. This prevents submitting broken or incomplete code and catches failures before they reach review.

Identify and run the proving commands for tests, builds, and lint checks specific to your project. Read the complete output and exit status for each command to confirm success, then only declare the task complete once all verification evidence confirms passing status.

Verification before completion catches incomplete work, broken builds, and failing tests locally, eliminating rework cycles and rejected PRs. It ensures honest, evidence-based reporting of code quality and prevents CI pipeline failures that waste team time.

Yes, the skill covers CI pipelines, local verification, and PR validations. It enforces the same principle across all environments: run full verification commands, read complete output, confirm the result matches your claim, then proceed with confidence.

Skipping verification leads to premature success claims, hidden defects, and wasted rework. This skill prevents that by enforcing mandatory verification: no completion claim is valid without fresh, explicit evidence from running tests, builds, and lint checks.

Verification-before-completion requires reading complete output and exit status to confirm every check passes before any success statement. Single test runs without evidence review allow false positives; this approach eliminates assumptions through mandatory, evidence-based validation.