Skip to main content
Connecting your AI to Tokenomist is simple. Point any MCP-compatible client at our hosted endpoint. Most MCP clients — Claude Desktop, Claude Code, Cursor, ChatGPT, Gemini CLI, and others — can be configured with a single JSON snippet.
Make sure your device has node installed. You can download it from nodejs.org/download.

Server info

Tokenomist MCP Server (Production)
  • Endpoint: https://mcp.tokenomist.ai/
  • Status: Stable — monitored, multi-replica deployment with 24/7 uptime.
  • Transport: MCP Streamable HTTP.
  • Auth: OAuth 2.1 with PKCE. A Tokenomist API key is required (used once during sign-in). See Authentication.
  • Best for: Production AI agents and integrations, research assistants and chatbots that need tokenomics context, and quick prototyping against real Tokenomist data.

Configuration

Add the following to your client’s MCP config file (e.g. mcp_config.json or claude_desktop_config.json):
{
  "mcpServers": {
    "tokenomist": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp.tokenomist.ai/"
      ]
    }
  }
}
Where is my config file? The file name and location depend on your client. See modelcontextprotocol.io/quickstart for client-specific paths.

Connect your client

Free users must use the Claude Desktop app and modify the configuration file.
1

Locate the config file

Find claude_desktop_config.json on your system by following the instructions here.
2

Add the server config

Add the server configuration shown above.
3

Restart Claude Desktop

Restart the app for the changes to take effect.
4

Authorize

On first use, a browser tab opens to the Tokenomist sign-in page — paste your Tokenomist API key to authorize.
Once connected, head to Tools & prompts to see what your agent can do.