Can I use this with custom build scripts?

Yes, you can configure project-specific test commands in your CLAUDE.md or .claude/rules/ files, which the skill will prioritize during its detection phase.

Does this skill actually write code or just run tests?

It does both. It analyzes test failures to apply minimal code fixes and identifies refactoring opportunities to improve code quality once tests are green.

How do I prevent the skill from looping indefinitely?

You can use the --max-cycles argument to limit iterations, and the skill includes built-in auto-stop conditions if it detects it is stuck or failing the same test repeatedly.

Which testing frameworks does the TDD Test Loop support?

It supports major frameworks out-of-the-box, including npm/vitest for Node.js, pytest for Python, cargo test for Rust, and go test for Go. It can also detect custom commands in Makefiles.

TDD Test Loop

Name: TDD Test Loop
Author: laurigates

bylaurigates

•

Security & Testing

Automates the test-fix-refactor cycle to achieve passing tests and clean code using a Test-Driven Development workflow.

The TDD Test Loop skill streamlines the development process by managing a continuous Test-Driven Development (TDD) cycle within Claude. It automatically detects your project's testing framework—supporting Node.js, Python, Rust, Go, and more—executes test suites, diagnoses failures to implement minimal fixes, and identifies refactoring opportunities once tests pass. This skill is ideal for developers who want to maintain high code quality and test coverage while automating the repetitive feedback loop of running tests and fixing small implementation errors.

Key Features

01Automated refactoring phase to eliminate code duplication and improve clarity

02Intelligent failure analysis to diagnose root causes and implement minimal fixes

03Automatic test command detection for major languages and frameworks

04Customizable focus using test patterns and max-cycle iteration limits

05Detailed reporting of cycles, applied fixes, and performed refactorings

0610 GitHub stars

Use Cases

01Rapidly fixing regressions in a codebase after making architectural changes

02Automating the 'Green' and 'Refactor' phases of a new feature's TDD cycle

03Cleaning up complex logic and magic numbers while ensuring functional parity

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add laurigates/claude-plugins project-test-loop

For use in Claude.ai and ChatGPT

Download Skill