Can I use 'should work' or 'probably fixed' with this skill?

No. The skill identifies these as red flags. You must provide specific evidence, such as exit codes or test pass counts, rather than expressing subjective satisfaction.

Why does this skill forbid trusting agent success reports?

To ensure maximum reliability, the skill treats agent reports as claims that require independent verification through VCS diffs and fresh command execution before they are accepted.

How does the 'Gate Function' work?

The Gate Function is a 5-step process: Identify the proof command, Run the full command, Read the output/exit code, Verify the claim against the output, and only then make the claim.

What is the core principle of this skill?

The core principle is 'Evidence before assertions.' It requires running actual commands and confirming their output before claiming any work is fixed, passing, or complete.

Verification Before Completion

Name: Verification Before Completion
Author: monmacllcapp

bymonmacllcapp

0•

Seguridad y Pruebas

Enforces rigorous verification protocols and evidence-based reporting before any task or code change is claimed as complete.

This skill establishes an uncompromising 'Evidence Before Claims' protocol for AI-driven development, ensuring that no work is marked as finished without empirical proof. It prevents premature success assertions by requiring the execution of fresh verification commands—such as test suites, linters, and build scripts—before committing code or opening pull requests. By implementing a strict 'Gate Function,' the skill eliminates common developer pitfalls like 'it should work' or relying on outdated results, ultimately maintaining a high standard of honesty and reliability in the software development lifecycle.

Características Principales

01Implements a strict Gate Function to confirm output against intended results

020 GitHub stars

03Prevents rationalizations and vague claims like 'should work' or 'seems correct'

04Enforces independent verification of changes when delegating to other agents

05Supports Red-Green-Refactor patterns for verified regression testing

06Mandates fresh execution of verification commands before any success claim

Casos de Uso

01Confirming build stability and requirement fulfillment during complex task migrations

02Ensuring 100% test pass rates and clean linting before creating a pull request

03Validating bug fixes by confirming the failure state before verifying the resolution

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add monmacllcapp/skill-forks verification-before-completion

For use in Claude.ai and ChatGPT

Download Skill