For Claude organization accounts, setup is required only once by an Owner. All organization users can then access Enigma.

Prerequisites: Ensure you're logged in to the Enigma Console. If you don't have an account, register here.

Installation Steps:

1. Open Claude app → User menu (bottom left) → Settings
2. Navigate to Connectors → scroll down → Add Custom Connector
3. Enter these values:
   Name: Enigma Remote MCP Server URL: https://mcp.enigma.com/http
4. Click Add then Connect
5. Authenticate in browser using your Enigma Console credentials
6. Close authentication window when complete

Quick installation using the one-click button below, or manual configuration in settings.json.

Prerequisites: Ensure you're logged in to the Enigma Console. If you don't have an account, register here.

Manual configuration in settings.json:

{
  "mcpServers": {
    "enigma": {
      "url": "https://mcp.enigma.com/http"
    }
  }
}

Connect Enigma's business intelligence tools to ChatGPT and OpenAI-compatible platforms.

Prerequisites: Ensure you're logged in to the Enigma Console. If you don't have an account, register here.

Installation Steps:

1. Open ChatGPT → Settings → Beta Features
2. Enable Model Context Protocol (MCP)
3. Navigate to MCP Connections → Add Connection
4. Configure connection:
   Name: Enigma Server URL: https://mcp.enigma.com/http
5. Click Connect and authenticate with Enigma Console
6. Start using Enigma tools in your ChatGPT conversations

Build custom AI agents with direct MCP integration. Works with any framework (OpenAI Agents, LangChain, CrewAI, AutoGen, etc.).

Quick Integration:

# 1. Connect to MCP with HTTP-key
headers = {
    "x-api-key": "your_enigma_api_key",
    "Content-Type": "application/json"
}

response = requests.post(
    "https://mcp.enigma.com/http-key",
    json={"jsonrpc": "2.0", "method": "initialize", "id": 1}
)

# 2. Use session ID for all subsequent calls
headers["mcp-session-id"] = response.headers.get("mcp-session-id")

# 3. Wrap tools for your framework (example: OpenAI Agents)
@function_tool
def search_business(query: str):
    return requests.post(url, json={
        "method": "tools/call",
        "params": {"name": "search_business", "arguments": {"query": query}}
    }, headers=headers).json()["result"]