Setup

Connect LeadMagic MCP to Cursor, VS Code, Claude, Windsurf, Zed, Gemini, and 7 more AI clients

Connect the LeadMagic MCP server to your AI assistant in minutes. You’ll need your API key from app.leadmagic.io/settings/api.

The LeadMagic MCP server uses Streamable HTTP transport (MCP 2025 spec) at https://mcp.leadmagic.io/mcp. Your API key is passed via the x-leadmagic-key header on every request and is never stored server-side.

Prerequisites

1. Get your API key

   Sign up or log in at app.leadmagic.io and copy your API key from Settings > API.

2. Choose your client

   Follow the instructions for your AI assistant below.

Client Setup

Cursor

Cursor

Add the LeadMagic server to your Cursor MCP configuration.

File: .cursor/mcp.json (project-level) or ~/.cursor/mcp.json (global)

{
  "mcpServers": {
    "leadmagic": {
      "url": "https://mcp.leadmagic.io/mcp",
      "headers": {
        "x-leadmagic-key": "YOUR_API_KEY"
      }
    }
  }
}

1. Create or edit the config file

   Open .cursor/mcp.json in your project root (or ~/.cursor/mcp.json for all projects).

2. Paste the config above

   Replace YOUR_API_KEY with your LeadMagic API key.

3. Restart Cursor

   Restart Cursor for the MCP server to be detected.

Use project-level config (.cursor/mcp.json) to keep the key scoped. Use global config (~/.cursor/mcp.json) if you want LeadMagic available in every project.

VS Code

VS Code

Add the LeadMagic server to your VS Code MCP configuration.

File: .vscode/mcp.json (workspace) or user-level mcp.json

{
  "servers": {
    "leadmagic": {
      "type": "http",
      "url": "https://mcp.leadmagic.io/mcp",
      "headers": {
        "x-leadmagic-key": "YOUR_API_KEY"
      }
    }
  }
}

1. Create or edit the config file

   Open .vscode/mcp.json in your workspace root.

2. Paste the config above

   Replace YOUR_API_KEY with your LeadMagic API key.

3. Reload VS Code

   Run the Developer: Reload Window command (Cmd+Shift+P or Ctrl+Shift+P).

For secure key storage, use VS Code input variables:

{
  "inputs": [
    {
      "type": "promptString",
      "id": "leadmagic-key",
      "description": "LeadMagic API Key",
      "password": true
    }
  ],
  "servers": {
    "leadmagic": {
      "type": "http",
      "url": "https://mcp.leadmagic.io/mcp",
      "headers": {
        "x-leadmagic-key": "${input:leadmagic-key}"
      }
    }
  }
}

The password: true input variable prompts you for the key on each session start, so it never touches disk.

Claude Desktop

Claude Desktop

Claude Desktop does not natively support remote HTTP MCP servers with custom headers. Use the mcp-remote proxy to bridge the connection.

File: claude_desktop_config.json

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json
• Linux: ~/.config/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "leadmagic": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp.leadmagic.io/mcp",
        "--header",
        "x-leadmagic-key:YOUR_API_KEY"
      ]
    }
  }
}

1. Ensure Node.js is installed

   mcp-remote runs via npx, which requires Node.js 18+.

2. Edit the config file

   Open the config file at the path above for your OS.

3. Paste the config

   Replace YOUR_API_KEY with your LeadMagic API key.

4. Restart Claude Desktop

   Fully quit and reopen Claude Desktop.

The mcp-remote package is a community tool. If you encounter issues, check the mcp-remote documentation for updates.

Claude Code

Claude Code

Add the LeadMagic server with a single CLI command.

claude mcp add --transport http \
  --header "x-leadmagic-key:YOUR_API_KEY" \
  leadmagic https://mcp.leadmagic.io/mcp

Replace YOUR_API_KEY with your LeadMagic API key.

Scope options:

Flag	Scope	Config File
--scope local (default)	Current project only	~/.claude.json
--scope project	Shared with team	.mcp.json in project root
--scope user	All projects	~/.claude.json

Alternative — add via JSON:

claude mcp add-json leadmagic '{
  "type": "http",
  "url": "https://mcp.leadmagic.io/mcp",
  "headers": {
    "x-leadmagic-key": "YOUR_API_KEY"
  }
}'

Use --scope project to share the config with your team via .mcp.json in your repo. Team members will need their own API keys.

Windsurf

Windsurf (Codeium Cascade)

Add the LeadMagic server to your Windsurf MCP configuration.

File: ~/.codeium/mcp_config.json

{
  "mcpServers": {
    "leadmagic": {
      "serverUrl": "https://mcp.leadmagic.io/mcp",
      "headers": {
        "x-leadmagic-key": "YOUR_API_KEY"
      }
    }
  }
}

1. Open Windsurf settings

   Go to Settings > Tools > Windsurf Settings > Add Server > View Raw Config.

2. Paste the config above

   Replace YOUR_API_KEY with your LeadMagic API key.

3. Restart Windsurf

   Restart Windsurf for the server to be detected.

Windsurf supports env var interpolation. Use "${env:LEADMAGIC_API_KEY}" instead of hardcoding your key.

More Clients

The LeadMagic MCP server works with any client that supports Streamable HTTP transport. Here are setup instructions for 8 additional clients.

Zed

Zed uses mcp-remote to connect to remote HTTP servers with custom headers.

File: ~/.config/zed/settings.json

{
  "context_servers": {
    "leadmagic": {
      "source": "custom",
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp.leadmagic.io/mcp",
        "--header",
        "x-leadmagic-key:YOUR_API_KEY"
      ]
    }
  }
}

Or use Agent Panel > Settings > Add Custom Server and paste the command.

Requires Node.js 18+ for npx. Check the green indicator dot in the Agent Panel to confirm the server is active.

Cline

File: cline_mcp_settings.json (in VS Code extension storage)

Open the MCP Servers icon in Cline, then click Configure.

{
  "mcpServers": {
    "leadmagic": {
      "url": "https://mcp.leadmagic.io/mcp",
      "headers": {
        "x-leadmagic-key": "YOUR_API_KEY"
      },
      "disabled": false
    }
  }
}

Roo Code

File: .roo/mcp.json (project) or ~/.roo/mcp_settings.json (global)

{
  "mcpServers": {
    "leadmagic": {
      "url": "https://mcp.leadmagic.io/mcp",
      "headers": {
        "x-leadmagic-key": "YOUR_API_KEY"
      },
      "disabled": false
    }
  }
}

Project-level config (.roo/mcp.json) takes precedence over global.

OpenCode

File: opencode.json (project) or ~/.config/opencode/config.json (global)

{
  "mcp": {
    "leadmagic": {
      "type": "remote",
      "url": "https://mcp.leadmagic.io/mcp",
      "headers": {
        "x-leadmagic-key": "YOUR_API_KEY"
      },
      "enabled": true
    }
  }
}

Gemini CLI

Option 1 — CLI command:

gemini mcp add --transport http leadmagic https://mcp.leadmagic.io/mcp

Then edit ~/.gemini/settings.json to add the header:

Option 2 — Edit settings directly:

File: ~/.gemini/settings.json

{
  "mcpServers": {
    "leadmagic": {
      "httpUrl": "https://mcp.leadmagic.io/mcp",
      "headers": {
        "x-leadmagic-key": "YOUR_API_KEY"
      }
    }
  }
}

Gemini CLI uses httpUrl (not url) for Streamable HTTP servers.

JetBrains IDEs

IntelliJ IDEA, WebStorm, PyCharm, and other JetBrains IDEs (2025.1+) support MCP via the AI Assistant.

1. Go to Settings > AI Assistant > MCP
2. Click + to add a new server
3. Select HTTP transport
4. Enter the URL: https://mcp.leadmagic.io/mcp
5. Add header: x-leadmagic-key = YOUR_API_KEY
6. Restart the IDE

Amazon Q

File: .amazonq/default.json (project) or ~/.aws/amazonq/default.json (global)

1. Open the Q Developer Chat panel
2. Click the tools icon to open MCP configuration
3. Click + and select the scope (global or local)
4. Select http as the transport protocol
5. Enter the URL: https://mcp.leadmagic.io/mcp
6. Add header: x-leadmagic-key = YOUR_API_KEY
7. Save and configure tool permissions

Copilot Coding Agent (GitHub)

Configure MCP for the GitHub Copilot coding agent at the repository level.

1. Go to your GitHub repo Settings > Copilot > Coding agent
2. Add the MCP configuration:

{
  "mcpServers": {
    "leadmagic": {
      "type": "http",
      "url": "https://mcp.leadmagic.io/mcp",
      "headers": {
        "x-leadmagic-key": "YOUR_API_KEY"
      },
      "tools": ["*"]
    }
  }
}

Copilot coding agent does not support OAuth authentication or MCP resources/prompts — only tools. All 16 LeadMagic tools will work.

Any MCP-compatible client that supports Streamable HTTP transport will work with LeadMagic. If your client isn’t listed here, use the server URL https://mcp.leadmagic.io/mcp with the x-leadmagic-key header.

Verify the Connection

After setup, test the connection by asking your AI assistant:

“Check my LeadMagic credit balance”

If the server is connected, the AI will call the check_credits tool and return your balance. This tool is free and confirms that authentication is working.

Next Steps

Tool Reference

See all 16 tools with parameters and example prompts.

Authentication

API key details and security best practices.