About Build My Own
build-my-own is an AI-oriented CLI for recreating software from GitHub projects, aimed at developers who want to understand implementation principles while working with AI. It creates a project structure, prompts, and rules, and can also run as an stdio MCP server.
Key features
- 01Scaffold a rebuild from a GitHub project
- 02Generate prompts and rules for AI
- 03CLI and MCP server modes
- 04Explore implementation details with AI
Use cases
- 01Rebuild software based on a GitHub project
- 02Study an unfamiliar project's implementation with AI
- 03Start a project recreation in Cursor or Claude Code
MCP support
It can run as an stdio MCP server configured in Cursor, Claude Code, or other AI tools. In MCP mode, it makes its project-recreation workflow available to AI.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
1Saved tool list · updated 3w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| github_url | string | Yes | GitHub repository URL to clone (must end with .git) |
Details
From the project READMEbuild-my-own
Introduction
build-my-own is an AI-friendly CLI tool. Use build-my-own to replicate any GitHub project. It sets up the directory structure, prompts, and rules for AI, enabling it to assist you—more like guiding you—to recreate your desired project from scratch. Throughout the process, you’ll collaborate with AI to think through the details of project implementation and understand its principles. Ultimately, you’ll gain a fresh, deeper, and practical understanding of the project.
Prerequisites
- Node.js
gitcommand available
Usage
build-my-own offers two usage modes:
- MCP Server: Starts an Stdio MCP Server for use in Cursor, Claude Code, or other AI tools.
- CLI: Builds an AI-friendly directory structure and prompts via CLI.
MCP Server
Configure the MCP Server in your AI tool:
{
"mcpServers": {
"build-my-own": {
"command": "npx",
"args": ["-y", "build-my-own"]
}
}
}
Then, you can interact with AI. For example, to replicate the Redux project:
I want to rebuild Redux, help me with the build-my-own tool.
CLI
Install:
npm i -g build-my-own
Usage:
// github-project-url should be ended with `.git`
build-my-own --github <github-project-url>
Why I Built This Project
AI can generate code, but we humans need to master the code it produces. Here are two reasons why:
- Humans maintain the code: If we don’t understand the code AI generates, we can’t maintain it. If AI becomes unreliable or unable to fix issues it creates, the entire project could collapse.
- If you’re using AI to generate code you don’t understand, you’re just creating another potentially unstable product for the world, and you’re not growing personally.
Using AI alongside human knowledge to boost efficiency is crucial. Equally important is leveraging AI as a powerful tool for self-evolution, quickly internalizing unfamiliar knowledge into our own capabilities.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.