Does this work with any programming language?

Yes, it is language-agnostic and relies on the test commands, file paths, and linting rules defined in your project's CLAUDE.md file.

Can I use this for existing codebases?

Absolutely. It is designed to work with existing source and test files, making it ideal for adding features or fixing bugs in established projects.

What happens if an agent fails a review during the pipeline?

The skill includes a built-in fix loop that re-dispatches agents with specific feedback, allowing up to three rounds of corrections before escalating to the user.

How does it ensure the 'Red' state is valid?

It verifies that tests compile but fail at runtime, specifically checking for 'default-value traps' to ensure the test actually exercises the intended logic before any code is written.

What is the difference between inline and pipeline modes?

Inline mode is a fast, single-agent loop for small changes like bug fixes, while Pipeline mode uses 7 specialized agents (Test Writer, Reviewer, etc.) for complex module development.

TDD Orchestrator

Name: TDD Orchestrator
Author: kelp

bykelp

•

安全与测试

Enforces strict red-green-refactor workflows and multi-agent pipelines for robust Test-Driven Development in Claude Code.

The TDD Orchestrator skill transforms Claude Code into a rigorous Test-Driven Development engine, ensuring every code change is backed by failing tests before implementation. It intelligently switches between a lightweight inline red-green-refactor loop for small fixes and a comprehensive 7-stage multi-agent pipeline for complex module development. By automating test writing, peer reviews, and verification gates, it prevents logic duplication and ensures high-quality, production-ready code while maintaining strict adherence to TDD principles.

主要功能

012 GitHub stars

02Intelligent routing between inline RGR and a full 7-agent pipeline

03Integrated commit management for clean, atomic code changes

04Automated multi-stage reviews including test-first verification

05Language-agnostic design that integrates with existing project test suites

06Strict enforcement of the Red state to prevent false-positive tests

使用场景

01Building a new software module from scratch with complex behaviors

02Refactoring existing codebases while maintaining continuous test coverage

03Fixing critical bugs by reproducing them with a failing test first

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add kelp/kelp-claude-plugins tdd-orchestrate

For use in Claude.ai and ChatGPT

主要功能

012 GitHub stars

02Intelligent routing between inline RGR and a full 7-agent pipeline

03Integrated commit management for clean, atomic code changes

04Automated multi-stage reviews including test-first verification

05Language-agnostic design that integrates with existing project test suites

06Strict enforcement of the Red state to prevent false-positive tests

使用场景

01Building a new software module from scratch with complex behaviors

02Refactoring existing codebases while maintaining continuous test coverage

03Fixing critical bugs by reproducing them with a failing test first

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add kelp/kelp-claude-plugins tdd-orchestrate

For use in Claude.ai and ChatGPT