What are the 'Red Flags' this skill monitors?

It blocks speculative phrases such as 'should work,' 'seems to,' or 'probably,' as well as any expression of satisfaction before verification commands are executed.

Does this skill work with automated test suites?

Yes, it is specifically designed to handle unit tests, integration tests, linters, and build commands, requiring a 0-failure exit code for success claims.

What is the primary purpose of this skill?

The skill ensures that Claude never claims a task is complete or fixed without running fresh verification commands and presenting the actual output as evidence.

How does it handle delegated tasks?

When an agent is used to perform a task, this skill mandates an independent check of the VCS diff and a manual verification run rather than trusting the agent's success report.

Verification Before Completion

Name: Verification Before Completion
Author: pur3v4d3r

bypur3v4d3r

•

Security & Testing

Enforces a strict evidence-first protocol by requiring verifiable command output before any task can be marked as finished or successful.

The Verification Before Completion skill implements a mandatory 'Gate Function' to prevent premature or unverified claims about code status. It mandates the execution of fresh verification commands—such as tests, builds, or linters—and requires a thorough analysis of the actual output before any success assertions are made or pull requests are created. By eliminating reliance on assumptions like 'should work' or trusting unverified agent reports, this skill ensures high-integrity development workflows, reduces regressions, and maintains professional accountability through documented evidence.

Key Features

01Evidence-based checklist for requirement compliance

02Prohibition of speculative language like 'should' or 'probably'

03Independent verification of delegated agent success reports

04Mandatory Gate Function workflow for all status claims

05Red-green TDD cycle verification for regression testing

061 GitHub stars

Use Cases

01Maintaining rigorous standards in multi-agent autonomous coding environments

02Preventing broken or uncompiled code from being committed to a repository

03Ensuring complex bug fixes are validated with fresh regression tests

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