Does this work with any programming language?

Yes, it is language-agnostic as it relies on the execution of shell commands, test runners, linters, and build tools specific to your project.

Why is this better than Claude's standard behavior?

While Claude is capable, it can sometimes be overconfident or assume code works based on logic alone. This skill enforces a 'Trust but Verify' approach with fresh, hard evidence.

How does this prevent regressions?

The skill emphasizes the Red-Green cycle, ensuring that tests are seen to fail before they are seen to pass, confirming the test is valid and the fix is effective.

What is the primary purpose of the Verification Before Completion skill?

It serves as a mandatory protocol that prevents the AI from claiming a task is done until it has successfully run a verification command and analyzed the actual output.

Strict Verification Gate

Name: Strict Verification Gate
Author: ironbob

byironbob

0•

Security & Testing

Enforces rigorous verification protocols to ensure all code claims are backed by real-time execution evidence before task completion.

The Strict Verification Gate skill implements an 'Iron Law' for Claude Code, preventing the AI from claiming tasks are complete or code is bug-free without first running and verifying the output of specific commands. It acts as a mandatory gate function, requiring fresh evidence from tests, builds, or linters before any success status is reported. By eliminating 'hallucinated' success reports and relying solely on current execution data, this skill ensures high-integrity development workflows and maintains trust between the AI agent and the human developer.

Key Features

01Comprehensive checklists for requirement and linter verification

02Evidence-based status reporting using real-time exit codes and logs

03Mandatory execution of verification commands before success claims

04Red-Green regression test validation to confirm bug fixes

05Prevention of speculative language like 'should work' or 'probably fixed'

060 GitHub stars

Use Cases

01Preventing false-positive test reports before committing code or creating PRs

02Ensuring complex bug fixes are verified against original symptoms using fresh data

03Maintaining rigorous quality standards during multi-step agent delegations

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

01Comprehensive checklists for requirement and linter verification

02Evidence-based status reporting using real-time exit codes and logs

03Mandatory execution of verification commands before success claims

04Red-Green regression test validation to confirm bug fixes

05Prevention of speculative language like 'should work' or 'probably fixed'

060 GitHub stars

Use Cases

01Preventing false-positive test reports before committing code or creating PRs

02Ensuring complex bug fixes are verified against original symptoms using fresh data

03Maintaining rigorous quality standards during multi-step agent delegations

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