verification-before-completion

Require fresh verification evidence before reporting completion or success.

115|54|Updated May 25, 2026
One-click install
npx skills add https://github.com/totvs/engpro-advpl-tlpp-skills --skill verification-before-completion-totvs
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: verification-before-completion
Source: https://github.com/totvs/engpro-advpl-tlpp-skills/tree/main/skills/superpowers/verification-before-completion
Command: npx skills add https://github.com/totvs/engpro-advpl-tlpp-skills --skill verification-before-completion-totvs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents false “done” or “fixed” claims by ensuring that completion status is backed by fresh verification evidence.

Core Features & Use Cases

  • Evidence-first completion rule: Requires running the actual proof command and reading its full output before any success or satisfaction statement.
  • Step-by-step gate function: Enforces identify → run the full command → read output/exit code → verify against the claim → then report.
  • Applies to engineering workflows: Covers tests, linting, builds, bug fixes, regressions, and readiness to commit or open PRs when communicating status.

Quick Start

Ask your AI agent to prepare and run the exact verification command for your change, show the complete output, and only then confirm whether the work is actually complete.

Frequently Asked Questions about verification-before-completion

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent false completion claims in software development before committing code?▼

Verify before completion requires running the actual proof command, reading its full output, and matching the reported state to the verification results before making any success claims. This enforces an evidence-first gate function for engineering tasks.

How do I verify code readiness before opening a pull request?▼

To verify code readiness before opening a pull request, execute the full proof command for your changes, review the complete output and exit code, and confirm the results match your completion claim before asserting the work is done.

What is the step-by-step process for build validation before reporting task status?▼

Build validation requires identifying the verification needed, running the full command, reading the output and exit code, verifying results against your claim, and only then reporting the status. This gate function prevents false fixed or done claims.

Does verification before completion work with CI pipelines and regression checks?▼

Yes, verification before completion applies to CI pipelines and regression checks by requiring fresh evidence from running tests, linting, or builds before any satisfaction statement. It covers standard engineering workflows including readiness to commit or open PRs.

Why does my AI agent claim a bug is fixed without showing test output?▼

AI agents may claim a bug is fixed without showing test output because they skip the verification gate. To stop this, require the agent to execute the actual proof command, display complete output, and match results to the claim before confirming completion.

When should I not use automated verification before claiming a task is complete?▼

You should not skip verification before claiming completion when a task involves tests, linting, builds, or regressions. If no proof command can be executed to generate fresh output evidence, the verification gate cannot validate the completion claim.