Run verification commands and confirm exit status before asserting completion.
npx skills add https://github.com/tta-lab/ttal-cli --skill sp-verify
Please help me install this Agent Skill. Skill: sp-verify Source: https://github.com/tta-lab/ttal-cli/tree/main/skills/sp-verify Command: npx skills add https://github.com/tta-lab/ttal-cli --skill sp-verify
It prevents false or premature “done/passing/complete” statements by forcing you to run the exact verification command and use fresh evidence before you assert success.
Use sp-verify before committing, pushing, or opening a PR to ensure your tests, build, or other required checks are freshly run and supported by the latest command output.
High-intent search queries and answers about installing and using this skill.
To prevent false completion claims in CI workflows, you must require fresh verification evidence by identifying the proving command, executing the full verification, and confirming exit status and output before asserting success.
Evidence-based verification for release readiness is the process of running exact proof commands and reading complete output, including exit codes and failure counts, to stop rationalizations like “should pass” or “looks correct” before claiming a task is complete.
To verify tests, linting, and builds before opening a pull request, you should run the specific proof command, check the complete output, and only declare passing status after confirming the exit code and failure counts from the latest run.
Yes, this verification approach works for bug fixes and regression cycles by enforcing an evidence-first completion gate that requires identifying the proof command, running the full check, and confirming exit status before asserting any positive status claim.
You need fresh verification evidence instead of trusting agent-reported success to establish safety guardrails against subtle failure modes, preventing premature or dishonest claims of completion when checks only appear correct without independent verification.
Asserting task completion without checking exit codes allows partial checks or rationalized success statements to bypass safety guardrails, risking false claims of tests, builds, or bug fixes passing when the actual verification output indicates failures.
archibateRequire fresh execution of exact verification commands before asserting completion.
carlos-algmsVerify work-completion claims with fresh command output and exit status.
CityBear3Delegate build, test, lint, diff review, and readability checks to an implementation-verifier agent.
bugrabilgeRequire executed verification commands before claiming task completion.
