mfranzon

mfranzon / tdd

Public

tdd skill for coding agents

128
5
100% credibility
Found Feb 21, 2026 at 30 stars 4x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
AI Summary

A coaching tool for an AI coding assistant that breaks down features into small steps and guides users through writing checks first, then code to pass them, and finally improvements, with pauses for review.

How It Works

1
💡 Discover the Guide

You hear about a friendly coach that helps build software features safely, step by tiny step with checks along the way.

2
🛠️ Add to Your Helper

You simply place this coach into your AI coding assistant's tools folder so it's ready to use.

3
📝 Jot Down Your Plan

You write a short note describing the feature you want to create, like a shopping list for your idea.

4
🚀 Kick Off the Adventure

You ask the coach to start with your plan, and it suggests small, easy pieces to build one at a time for your okay.

5
🔍 Review Test Suggestions

For each piece, the coach shows a simple check that should fail at first, and you look it over before trying it.

6
✅ Make It Pass and Polish

You get ideas for basic code to make the check work, try it out to see everything pass, then suggestions to make it nicer.

7
📋 Finish Strong

You repeat for all pieces, always in charge, with full checks each time to keep things solid.

🎉 Feature Complete!

Your feature is built with reliable checks, plus a recap of your wins and tips for what's next—feels great!

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 30 to 128 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is tdd?

This Claude Code TDD skill on GitHub turns AI-assisted coding into a strict Test-Driven Development coach. Feed it a markdown plan file via the `/tdd path/to/plan.md` command, and it decomposes features into tiny increments, then guides you through Red-Green-Refactor cycles—proposing failing tests and minimal code to temp files for your review before running pytest (Python), vitest (TypeScript), or Go's testing suite. It solves the problem of drifting from TDD discipline when using tools like GitHub Copilot, enforcing human-in-the-loop pauses so you stay in control while building focused, passing tests.

Why is it gaining traction?

Unlike loose AI code generators, this Claude Code TDD GitHub skill acts as a TDD guardrail, running full test suites at every green/refactor step and prioritizing simplest cases first—making TDD skills stick without overwhelming you. Developers dig the temp-file proposals that never touch your source, plus multi-language support for Python TDD GitHub workflows or TypeScript setups. It's a lightweight TDD agent skill that hooks solo coders practicing TDD katas by blending AI smarts with your judgment.

Who should use this?

Backend engineers on Python or Go projects needing TDD enforcement during feature spikes. TypeScript devs building libraries who want AI-guided red-green cycles without Copilot's freewheeling suggestions. Solo practitioners running TDD skills drills or katas in Claude Code sessions.

Verdict

Worth a quick install for Claude Code users serious about TDD—copy to your skills dir and test on a small plan—but its 1.0% credibility score and 18 stars reflect early-stage maturity with just a solid README, no tests or examples. Try it for Python TDD GitHub experiments before committing to production flows.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.