What the tech

Search

/ opens search from anywhere. Esc closes it.

Tools

Get the newsletter

tdd: a Claude Code skill by Matt Pocock

For developers who want Claude to build features test first, one failing test at a time. Skip it if you want a whole test suite written in one go.

Score
87of 100
Rank
#18of 239
Installs
1.2Mon skills.sh
Security
Clear 5 scanners
Updated
2026
Context
~38tokens always

Install

npx skills add https://github.com/mattpocock/skills --skill tdd

Run it in your terminal. View the source

Using the Claude app rather than Claude Code? You can add a skill there too. How to add it in the app

What does tdd do?

tdd is a Claude skill that sets the rules for a red to green loop: write one failing test, then only enough code to pass it, then repeat. Before any test is written, Claude must list the seams, meaning the public interfaces to test at, note what each one catches and misses, and get your agreement. It reads GLOSSARY.md if present so test names match your domain words, and it respects ADRs. It warns against tests tied to internals, tautological assertions and writing all tests before any code. Refactoring stays out of the loop and goes to a separate code-review skill. Examples live in tests.md and mocking.md.

When should you use tdd?

Use it when

  • Building a new feature test first
  • Fixing a bug by first writing a test that fails
  • Adding integration tests at agreed public interfaces

Skip it when

  • You want code written without tests
  • You need many tests generated for existing code at once

What it needs and what it can touch

What it can do

  • Reads and edits project files

From reading its instructions, not from running it.

Why it scores 87

Strongest on upkeep (100 of 100), weakest on trust (65).

  • Adoption35 points 93
  • Upkeep20 points 100
  • Trust20 points 65
  • Quality15 points 85
  • Reader answers10 points Not enough answers yet, so this does not count

How the formula works

Is tdd safe to install?

Every scanner that checked it passed it. Scanners often miss what other scanners catch, and a skill can change after a scan. Read its source before you install it.

ScannerResultDate
Gen Agent Trust HubPassedOct 7, 2026
SocketPassedOct 7, 2026
SnykLow riskOct 7, 2026
RunlayerPassedFeb 26, 2026
ZeroLeaksPassedApr 16, 2026

Details

Installs
1,228,170
Repository stars
321,400
Last repository change
Tokens always loaded
about 38
Tokens when used
about 859
License
MIT
Data confidence
High

Alternatives to tdd

All skills for this job