.mcp.json Configuration Guide

What is Model Context Protocol?

Model Context Protocol (MCP) is an open standard for connecting AI applications to external data sources and tools. Think of it like USB-C for AI — a universal way to plug capabilities into LLM applications regardless of which AI assistant you're using.

While CLAUDE.md and .cursorrules tell AI how to behave, MCP gives AI what it can access — databases, APIs, file systems, web services, and custom tools.

Configuration File Locations

• Cursor (project) — .cursor/mcp.json
• Cursor (global) — ~/.cursor/mcp.json
• VS Code — .vscode/mcp.json
• OpenAI Codex — ~/.codex/config.toml

.mcp.json Format

MCP configuration uses JSON format to define servers and their capabilities:

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/Users/me/projects"],
      "env": {}
    },
    "github": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github"],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_your_token_here"
      }
    },
    "postgres": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-postgres"],
      "env": {
        "DATABASE_URL": "postgresql://user:pass@localhost:5432/mydb"
      }
    },
    "custom-api": {
      "command": "node",
      "args": ["./mcp-servers/my-api-server.js"],
      "env": {
        "API_KEY": "your_api_key"
      }
    }
  }
}

Configuration Fields

• command — The executable to run (npx, node, python, etc.)
• args — Arguments passed to the command
• env — Environment variables for the server process

Transport Types

MCP supports three transport types:

Popular MCP Servers

Related Guides

Frequently Asked Questions

Manage Your .mcp.json Files Visually

ClaudeMDEditor lets you browse and organize MCP configurations alongside CLAUDE.md, .cursorrules, and all your other AI coding assistant config files.

One-time purchase for macOS Tahoe (26+)