Can this skill be used for regression testing?

Yes, it supports strict TDD patterns where the AI must prove a test fails when the fix is reverted and passes when the fix is applied (Red-Green cycle).

How does it handle speculative language like 'should work'?

The skill treats words like 'should', 'probably', and 'seems to' as red flags. It triggers a stop-and-verify protocol that requires the AI to run commands before proceeding.

What is the 'Iron Law' of this skill?

The Iron Law states: 'No completion claims without fresh verification evidence.' It prevents the AI from assuming code works without running the relevant commands in the current session.

Why is this better than standard AI task completion?

It removes 'rationalization' where an AI might think code is correct because it looks right. This skill forces external validation via the terminal, ensuring the code actually executes correctly.

Verification Before Completion

Name: Verification Before Completion
Author: pur3v4d3r

bypur3v4d3r

•

Security & Testing

Enforces a strict evidence-based workflow requiring real-time command execution and output verification before any task is marked as complete.

This skill serves as a rigorous quality gate for Claude Code, mandating that all success claims, status updates, and Pull Request readiness be backed by fresh, objective evidence. It implements an 'Iron Law' of verification that prevents premature assertions, 'hallucinated' successes, and the common pitfalls of rationalization. By following a structured Identify-Run-Read-Verify protocol, it ensures that tests, builds, and linting processes are actually executed and confirmed through exit codes and logs before the AI reports completion to the user.

Key Features

01Red-Green TDD verification patterns for regression testing

02Automated detection of speculative language and red-flag phrases

03Independent verification of delegated agent results and VCS diffs

04Mandatory Identify-Run-Read-Verify gate for all status claims

05Evidence-based requirement tracking against implementation checklists

061 GitHub stars

Use Cases

01Confirming a full project build succeeds after major refactoring before creating a PR

02Ensuring a bug is truly fixed by verifying original symptoms pass after a code change

03Validating that all requirements are met by cross-referencing logs with original project specs

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add pur3v4d3r/pur3-pkb-codebase verification-before-completion

For use in Claude.ai and ChatGPT

Download Skill