What are the 'Red Flags' this skill monitors?

It flags phrases like 'should work,' 'I'm confident,' or 'seems to,' requiring the AI to replace assumptions with verified command results and exit codes.

Why is evidence-based verification important for AI coding?

AI agents can sometimes hallucinate success or overlook subtle failures; requiring terminal output evidence ensures that the work is actually functional and meets requirements before proceeding.

How does the 'Gate Function' work?

The Gate Function requires the AI to identify a proof command, run it, read the full output, and only then make a status claim based on that specific evidence.

Does this skill support Playwright or UI testing?

Yes, it specifically mandates running tools like Playwright in the terminal for full visibility rather than in the background to ensure the AI actually sees and confirms the results.

Verify Before Complete

Name: Verify Before Complete
Author: RockwallJMC

byRockwallJMC

0•

安全与测试

Enforces a strict evidence-based workflow that requires running fresh verification commands before claiming any task is finished or successful.

This skill implements an 'Iron Law' of verification for Claude Code, preventing premature, assumed, or dishonest completion claims during the development process. It establishes a mandatory gate function where the AI must identify a proof command, execute it in a visible terminal, and analyze the full output before asserting that a bug is fixed, a build passes, or requirements are met. By eliminating vague language like 'should' or 'probably,' it ensures a high-integrity development cycle rooted in empirical evidence rather than confidence or partial checks.

主要功能

01Visibility-first terminal execution for transparent command monitoring

02Strict language enforcement to eliminate assumptions and vague assertions

03Regression test verification using TDD Red-Green validation patterns

04Mandatory fresh verification evidence before any success claims

05Standardized Gate Function for analyzing exit codes and failure counts

060 GitHub stars

使用场景

01Ensuring full build and lint compliance before committing code or opening PRs

02Validating bug fixes by verifying the complete failure-to-success cycle

03Verifying agent-delegated tasks through VCS diff analysis and independent checks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add rockwalljmc/threadbilt-v0 VERIFY-BEFORE-COMPLETE

For use in Claude.ai and ChatGPT

Download Skill