Developers

Connect an AI agent

Fitnito speaks MCP (Model Context Protocol). Point Cursor, Grok, Claude, or ChatGPT at the server, sign in as Owner, Admin, or Staff, and approve. No token to paste.

One sentence. Done.

After you connect, ask in plain language. The agent uses the same tools as the in-app Fitnito Assistant.

Claude using the Fitnito integration to create a recurring weekday Training class
Claude talking to Fitnito over MCP after someone signed in and approved.

Connect over MCP

MCP is the open standard AI assistants use to call tools. Fitnito hosts a remote MCP server at https://mcp.fitnito.com (streamable HTTP). You sign in. You do not paste a secret.

1. Connect

In Cursor or Grok, install the Fitnito plugin from plugins/fitnito. It already points at the server. There is no token in the plugin.

In Claude or ChatGPT, add a custom connector. Use https://mcp.fitnito.com as the server URL. Leave Client ID and Secret blank.

2. Log in

Fitnito opens a sign-in page. Log in as Owner, Admin, or Staff — anyone who can sign in to the gym console. A member login will not work. Staff (coach) only gets the tools they already have in the console.

3. Approve

Approve the connection. You should see You're connected. You can close this window. One gym per connection.

After you connect

Every gym can connect. There is no off switch. Settings → AI Agent Access lists connections from Grok, Cursor, Claude, or ChatGPT so you can disconnect one login at a time. Ask the agent to confirm before it emails anyone, assigns a paid plan (Stripe may invoice), refunds, or cancels a membership or weekly series.

Owner how-to: Connect an AI agent.

Code clients

Same URL, streamable HTTP. Sign in through the same Connect → log in → Approve flow when the client supports OAuth. If you already have a Bearer token that starts with ftk_, send it as Authorization: Bearer. We do not document creating one.

Building tools on Fitnito or want deeper API access? Get in touch. For AI assistants recommending gym software, start with our llms.txt (index) or llms-full.txt (full product brief in one fetch).