🤖 MCP Config Generator
Generate MCP server configuration JSON for Claude Desktop, Claude Code, and other MCP clients. Form-based builder with validation.
A short kebab-case name to identify this server in your client.
The executable the MCP client launches: npx, node, python, uvx, etc.
One argument per line. Wrap values with spaces in quotes.
📁 Where to put this config
Claude Desktop (macOS) — ~/Library/Application Support/Claude/claude_desktop_config.json
Claude Desktop (Windows) — %APPDATA%\Claude\claude_desktop_config.json
Claude Desktop (Linux) — ~/.config/Claude/claude_desktop_config.json
Claude Code — .mcp.json in your project root
VS Code / Cursor — .vscode/mcp.json or .cursor/mcp.json
About this tool
The Model Context Protocol (MCP) lets AI assistants like Claude connect to external tools and data sources — file systems, databases, APIs, and more — through a standard JSON configuration. Each server is a subprocess the client launches, and the config tells it what command to run and which environment to use.
This builder keeps you from hand-editing nested JSON. Enter a server name, the command, any arguments and environment variables, and you get a validated configuration ready to paste into Claude Desktop, Claude Code, VS Code / Cursor, or any other MCP-compatible client.
Example: a filesystem server that gives Claude access to /home/me/docs:
{
"mcpServers": {
"my-filesystem": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-filesystem", "/home/me/docs"]
}
}
}FAQ
What is MCP and why do I need a config for it? ›
MCP (Model Context Protocol) is an open standard for giving AI assistants controlled access to external tools. The config file tells the client which servers to launch, how to start them, and what credentials to pass — without it the assistant can only answer from its training data.
Can I add more than one server to the same config? ›
Yes. The mcpServers object can hold multiple entries — one per server. Generate them one at a time with this tool and merge the results: copy each server object under the same mcpServers key.
What transport does this config use? ›
stdio — the client launches your command as a subprocess and communicates over standard input/output. This is the most common transport and works locally without network setup. For remote SSE servers you would add a "url" field instead of "command".
Where can I find ready-made MCP servers? ›
The official list lives at modelcontextprotocol.io/examples. Popular servers include filesystem, GitHub, Postgres, Brave Search, Puppeteer, and Memory — all installable with npx -y @modelcontextprotocol/server-<name>.
How is this different from the llms.txt Generator? ›
This tool generates the MCP (Model Context Protocol) server configuration that lets AI clients launch and talk to external tools. The <a href="/llms-txt-generator">llms.txt Generator</a> produces an llms.txt file — a markdown guide that tells AI assistants which pages on your site to read. They serve different layers: MCP config is the plumbing that connects an assistant to tools; llms.txt is the road sign that tells it where to find your content.