Loading...

CodeCrafters

Catalog Pricing

Build your own Claude Code

Learn about LLM APIs, tool calling, agent loops and more

C++

Java

JavaScript

Python

Rust

TypeScript

Zig

Build your own Claude Code

This challenge is free to try when it's in beta. We keep challenges in beta for a few weeks to gather feedback.

Claude Code is a terminal-based AI coding assistant that uses Large Language Models (LLMs) to understand code and complete programming tasks. In this challenge, you'll build your own Claude Code that is capable of editing files, running commands, and iterating until the task is done.

Along the way, you'll learn about LLM APIs, tool calling, agent loops, and how to integrate multiple tools into an AI assistant.

What exactly will I build?

What exactly will I learn?

Why should I build such a project?

What are the pre-requisites for this challenge?

Stages

Communicate with the LLM

Advertise the read tool

Execute the read tool

Implement the agent loop

Implement the write tool

Implement the bash tool

Skills

Advertise skills to the LLM

Invoke a skill by name

Pass arguments to a skill

Stack multiple skills

Run a script bundled with a skill

Let the model choose a skill

Run a skill in a subagent

RECENT ATTEMPTS 0

“ There are few sites I like as much that have a step by step guide. The real-time feedback is so good, it's creepy!

Ananthalakshmi Sankar Automation Engineer at Apple “ I think the instant feedback right there in the git push is really cool. Didn't even know that was possible!

Patrick Burris Senior Software Developer, CenturyLink