Why use this instead of standard linting?

Linting only checks syntax; this skill enforces a mindset of holistic verification including builds, tests, and requirement checklists before any success is claimed.

What is the 'Iron Law' in this skill?

The Iron Law dictates that no completion claims can be made without fresh, specific verification evidence derived from actual command outputs.

How does the Gate Function work?

It is a 5-step process: Identify the proof command, Run it fresh, Read the full output, Verify the claim against the output, and only then state the claim.

Does this skill work with automated agents?

Yes, it includes specific patterns to verify VCS diffs and independent results when work is delegated to other agents to prevent trust failures.

Verified Completion Workflow

Name: Verified Completion Workflow
Author: ironbob

byironbob

0•

Security & Testing

Enforces rigorous, evidence-based verification protocols before any task is marked as complete or committed to a repository.

This skill implements a strict 'Iron Law' for software development, preventing developers and AI agents from claiming success without fresh, verifiable evidence. By mandating a specific identification, execution, and reading sequence for verification commands (like tests, linters, and builds), it eliminates false positives, broken builds, and premature pull requests. It is designed to maintain high trust and code quality by replacing assumptions with raw command output and exit codes, ensuring that every 'Done' status is backed by actual data.

Key Features

010 GitHub stars

02Five-step Gate Function for identifying, running, and reading verification evidence

03Independent verification protocols for delegated AI agent tasks

04Mandatory verification command execution before PR or commit creation

05Strict Red-Green TDD verification patterns for regression testing

06Standardized checklists for requirements validation and build status

Use Cases

01Preventing 'should pass' claims in high-stakes production environments

02Auditing AI agent output to ensure proposed changes actually meet requirements

03Validating complex bug fixes by re-testing original symptoms systematically

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ironbob/superpower_light verification-before-completion

For use in Claude.ai and ChatGPT

Key Features

010 GitHub stars

02Five-step Gate Function for identifying, running, and reading verification evidence

03Independent verification protocols for delegated AI agent tasks

04Mandatory verification command execution before PR or commit creation

05Strict Red-Green TDD verification patterns for regression testing

06Standardized checklists for requirements validation and build status

Use Cases

01Preventing 'should pass' claims in high-stakes production environments

02Auditing AI agent output to ensure proposed changes actually meet requirements

03Validating complex bug fixes by re-testing original symptoms systematically

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ironbob/superpower_light verification-before-completion

For use in Claude.ai and ChatGPT