Best Projects Built With Claude Code, Codex, and Gemini CLI

What Builders Are Making With Claude Code, Codex, and Gemini CLI

A curated showcase of projects built with AI coding tools. Browse featured work or submit your own.

Three terminal tools now let developers build entire applications through conversation. Anthropic's Claude Code, OpenAI's Codex CLI, and Google's Gemini CLI each take a different approach to the same problem: turning plain-language instructions into working software. Some developers use them for quick prototypes. Others ship production code.

This page collects the most interesting projects we find. Some arrive through our inbox. Others surface on GitHub or in developer communities. We update the showcase regularly and welcome submissions through the form below.


Claude's C Compiler

by Anthropic

Claude Code

Sixteen Claude Opus 4 agents, running as Claude Code instances, autonomously wrote a complete C compiler in Rust from scratch over roughly 2,000 sessions. It targets x86-64, ARM, and RISC-V, and can compile a booting Linux 6.9 kernel along with PostgreSQL, SQLite, Redis, and FFmpeg.

View on GitHub

Pokemon Showdown RS

by Christopher Chedeau (vjeux)

Claude Code

A 1-to-1 port of the entire Pokemon Showdown battle simulator from TypeScript to Rust, done by Claude Code running autonomously for a month. The creator, known for Prettier and React Native, had never written Rust before. Over 5,000 commits achieved 99.997% behavioral parity across 2.4 million test seeds.

View on GitHub

Submit Your Project

Built something with an AI coding tool? We want to see it. Fill out the form below and we will review your submission for the showcase.

Thanks! We received your submission and will review it soon.

Your email is used only to follow up on your submission.


What We're Looking For

We feature projects that meet four criteria.

Original work. The project solves a real problem or explores a novel idea. Thin wrappers around existing APIs and tutorial follow-alongs do not qualify.

Public repository. The code lives on GitHub where anyone can inspect and run it.

It works. The project functions as described. We test every submission before featuring it.

Built with AI coding tools. The developer used Claude Code, Codex CLI, Gemini CLI, or a comparable agent during development. The tool must have played a meaningful role, not just autocompleted a few lines.

We review submissions within two weeks. Not every project gets featured, but we read them all.

Great! You’ve successfully signed up.

Welcome back! You've successfully signed in.

You've successfully subscribed to Implicator.ai.

Success! Check your email for magic link to sign-in.

Success! Your billing info has been updated.

Your billing was not updated.