What is the primary rule of this skill?

The core principle is 'Evidence before claims, always.' It prohibits declaring work finished or tests passing without running a fresh command and verifying the output immediately beforehand.

Does this skill work with TDD workflows?

Yes, it specifically enforces a Red-Green-Refactor cycle for regression tests, requiring the AI to prove a test fails without the fix before proving it passes with the fix.

How does this prevent 'false positives' in AI coding?

It identifies 'Red Flag' language such as 'should work' or 'seems to' and replaces it with a mandatory gate function: Identify, Run, Read, and Verify before making any status claim.

Why is this better than standard AI task completion?

Standard AI agents may hallucinate success or rely on partial checks. This skill creates a non-negotiable requirement for physical evidence (command exit codes and logs) to maintain trust and code stability.

Rigorous Completion Verification

Name: Rigorous Completion Verification
Author: januarfonti

byjanuarfonti

0•

보안 및 테스팅

Enforces a strict evidence-based protocol requiring fresh command output before any task is marked as complete or successful.

This skill acts as a critical quality gate for Claude Code, preventing premature 'success' claims by mandating that every assertion is backed by fresh, verifiable evidence. By implementing the 'Evidence Before Assertions' principle, it forces the AI to run tests, check linters, and verify build statuses before making any claims or creating pull requests. This prevents the common pitfall of assuming code works based on confidence alone, ensuring that only fully validated, working code is committed to your repository.

주요 기능

01Strict 'No Completion Without Evidence' enforcement

02Mandatory Red-Green cycle verification for regression tests

03Detection and prevention of success-based rationalizations

04Agent delegation validation via VCS diff checking

050 GitHub stars

06Comprehensive checklist verification for project requirements

사용 사례

01Pre-commit validation to ensure bug fixes actually solve the reported issue

02Ensuring CI/CD readiness by running full local build and lint suites

03Verifying the work of delegated sub-agents before accepting their changes

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add januarfonti/opencode-template-configured verification-before-completion

For use in Claude.ai and ChatGPT

Download Skill

주요 기능

01Strict 'No Completion Without Evidence' enforcement

02Mandatory Red-Green cycle verification for regression tests

03Detection and prevention of success-based rationalizations

04Agent delegation validation via VCS diff checking

050 GitHub stars

06Comprehensive checklist verification for project requirements

사용 사례

01Pre-commit validation to ensure bug fixes actually solve the reported issue

02Ensuring CI/CD readiness by running full local build and lint suites

03Verifying the work of delegated sub-agents before accepting their changes

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add januarfonti/opencode-template-configured verification-before-completion

For use in Claude.ai and ChatGPT

Download Skill