A model-invoked red-green-refactor loop: the agent writes a failing test first, makes it pass with the smallest change, then refactors, and refuses to skip steps.
skillgild install tdd
License: MIT · Created by Matt Pocock
Pick your agent. The wrapper only tells your agent how to start a session. The skill's instructions arrive with each session, after your access is checked.
Sign in once from your terminal.
skillgild loginInstall this skill for Claude Code.
skillgild install tdd --agent claude-codeRegister the SkillGild MCP server with Claude Code.
claude mcp add --scope user skillgild -- $(command -v skillgild) mcpStart a session, follow its guide, then end it. Create an API key under Account and pass it as a bearer token.
# Prints the session ID, its tools and the instructions for your agent
skillgild session tdd
skillgild end <session-id>Run skillgild install tdd --agent claude-code and connect the SkillGild MCP server (skillgild mcp). When a task matches, your agent starts a session, which uses one of your 25 free runs this month, and follows the skill's instructions. Use --agent codex, cursor or gemini-cli for other clients.
Yes. It includes 25 free runs every month on SkillGild. TDD (Matt Pocock) is open source (MIT) by Matt Pocock; SkillGild packages it from https://github.com/mattpocock/skills at a pinned commit and credits the author.
It runs in your coding agent with your own model. SkillGild checks your monthly allowance and delivers the skill's instructions for each session; it makes no model call and never sees your project unless your agent sends it.
The project lists Claude Code, Codex and other Agent Skills hosts. It is a plain Agent Skills folder (SKILL.md plus supporting files), so any client that reads that format can load it.
Setup steps and worked examples for this kind of skill.
Karpathy's observations on how LLMs fail at coding, codified into four rules the agent follows: think before coding (state assumptions, ask instead of guessing), simplicity first, surgical changes, and goal-driven execution.
A large community collection: 380+ skills, 30+ agents and 70+ commands for Claude Code, Codex, Gemini CLI, Cursor and other agents, organised by job (engineering, marketing, product, compliance, C-level advisory, research, finance, operations, daily productivity).
Anthropic's official code-review plugin for Claude Code: runs a structured review of a diff or pull request for correctness, standards and spec adherence and reports findings with locations and severity.
Anthropic's official plugin that adds commit, push and pull-request commands to Claude Code: /commit-commands:commit writes a conventional message from the diff, pushes, and can open the PR.
Makes the agent interview you about a plan or design, one question at a time with a recommended answer attached, until every branch of the decision tree is resolved and you both share the same understanding.
A context-engineering framework that runs Claude Code through a disciplined phase loop (new-project or onboard, plan-phase, execute-phase in parallel waves, verify-work, ship) with research and execution in fresh-context subagents so the main session never rots.
Compacts a long conversation into a handoff document another agent (or a fresh session) can continue from: the goal, decisions made, current state, open questions and the exact next step.
Anthropic's skill for building MCP servers: it walks the agent through tool design, schemas, transports and testing so the server it generates follows the Model Context Protocol correctly and works in Claude Code, Cursor and Codex.
Turn a feature request into an OpenSpec change before any code is written: proposal, delta specs, design and implementation tasks, grounded in the project's existing specs.
The lazy senior developer ruleset: before writing code the agent climbs a decision ladder (is it needed, does something reusable exist, is there a built-in) and only then writes the minimum.