What is the 'Iron Law' in this skill?

The Iron Law states: 'No completion claims without fresh verification evidence.' You cannot claim a task is passed unless the verification command was run in the current session.

Does this skill work with existing CI/CD tools?

Yes, it is designed to complement existing tools by ensuring the developer/agent confirms linter, test, and build success locally before pushing to CI.

What happens if a verification command fails?

The protocol requires stating the actual status with evidence of the failure, preventing the common mistake of ignoring 'small' errors or partial failures.

Can I use this for non-coding tasks?

Absolutely. The checklist and evidence-based verification logic can be applied to documentation, requirement gathering, and architectural planning.

How does this prevent AI hallucinations regarding code status?

It forces the AI to identify a specific command that proves a claim, execute it, and read the full output before expressing satisfaction or moving to the next task.

Verification Before Completion

Name: Verification Before Completion
Author: SwervoJake

bySwervoJake

0•

Security & Testing

Enforces a mandatory verification protocol requiring objective evidence before any task is claimed as complete or successful.

This skill implements a strict 'Evidence Before Claims' protocol for Claude Code, preventing premature assertions of success or false progress reports. It acts as a rigorous gatekeeper that requires the execution of fresh verification commands—such as tests, builds, and linters—before any code is committed, pushed, or marked as done. By mandating the inspection of actual exit codes and terminal outputs, it eliminates the risk of 'hallucinated' fixes and ensures that every claim of completion is backed by reproducible, objective proof of correctness.

Key Features

01Comprehensive validation of build status and requirement checklists

02Prevention of common rationalization traps like 'it should work'

030 GitHub stars

04Evidence-based reporting using fresh, live command outputs

05Strict Red-Green cycle enforcement for regression testing

06Mandatory 'Gate Function' workflow before any success claims

Use Cases

01Ensuring bug fixes are verified through specific regression tests

02Validating full requirement fulfillment before project handoff

03Preventing broken code from being committed to a repository

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add swervojake/claaaude-studio verification-before-completion

For use in Claude.ai and ChatGPT

Key Features

01Comprehensive validation of build status and requirement checklists

02Prevention of common rationalization traps like 'it should work'

030 GitHub stars

04Evidence-based reporting using fresh, live command outputs

05Strict Red-Green cycle enforcement for regression testing

06Mandatory 'Gate Function' workflow before any success claims

Use Cases

01Ensuring bug fixes are verified through specific regression tests

02Validating full requirement fulfillment before project handoff

03Preventing broken code from being committed to a repository

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add swervojake/claaaude-studio verification-before-completion

For use in Claude.ai and ChatGPT