MCP Server Quickstart

Get an API key

MCP auth uses the same lb_api_ Bearer token as the REST API. Follow the Auth guide to create a key with at least the read and write scopes, then come back here.

Configure your MCP client

Add the ListenBrief server to your client's config file. Below is the snippet for Claude Desktop (~/.config/claude/claude_desktop_config.json on Linux/macOS, %APPDATA%\Claude\claude_desktop_config.json on Windows). Restart Claude Desktop after saving.

{
  "mcpServers": {
    "listenbrief": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://listenbrief.com/mcp/v1",
        "--header",
        "Authorization: Bearer *** YOUR_LB_API_KEY_HERE"
      ]
    }
  }
}

# Verify the server is reachable and returns its tool list
curl -X POST https://listenbrief.com/mcp/v1 \
  -H "Authorization: Bearer lb_api_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/list",
    "params": {}
  }'

Discover available tools

Call tools/list to see all tools the server exposes. You can do this from any MCP client or directly with curl:

curl -X POST https://listenbrief.com/mcp/v1 \
  -H "Authorization: Bearer lb_api_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/list",
    "params": {}
  }'

The server exposes these 10 tools:

Call a tool

Use tools/call to invoke any tool by name. The example below creates a new briefing profile named "Morning Tech" and returns the resulting profile object.

Request:

curl -X POST https://listenbrief.com/mcp/v1 \
  -H "Authorization: Bearer lb_api_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "id": 2,
    "method": "tools/call",
    "params": {
      "name": "create_briefing_profile",
      "arguments": {
        "name": "Morning Tech",
        "interests": ["AI", "startups", "open source"],
        "length_minutes": 7
      }
    }
  }'

Response:

{
  "jsonrpc": "2.0",
  "id": 2,
  "result": {
    "content": [
      {
        "type": "text",
        "text": "{\"id\":\"prof_01abc...\",\"name\":\"Morning Tech\",\"interests\":[\"AI\",\"startups\",\"open source\"],\"length_minutes\":7,\"created_at\":\"2024-01-15T10:00:00Z\"}"
      }
    ],
    "isError": false
  }
}

Parse the content[0].text string as JSON to get the profile object. The id value (prof_01abc...) is what you'll pass to add_source, generate_briefing, and schedule_briefing.