Connect an AI assistant

Melissa works with AI assistants that support the Model Context Protocol (MCP). Connect your account once and ask about clients, appointments, notes and finances in plain language.

ChatGPT
Claude
Gemini
Copilot
Cursor

Server address

Add this address in your AI app as a custom connector (MCP server):

https://api.melissatreat.dev/v1/mcp

No API keys are needed: you sign in to Melissa and approve access in the browser.

How it works

Add the server address in your AI app.
Sign in to Melissa in the window that opens and review what the app will be able to do.
Click “Allow” and ask the assistant, for example: “Show my appointments this week”.

Step-by-step instructions

Claude (web and desktop)

Open Settings → Connectors and click “Add custom connector”.
Enter the name Melissa Treat, paste the server address and click “Add”.
Click “Connect”, sign in to Melissa and allow access.
In a chat, turn on Melissa Treat in the tools menu and ask your question.

Custom connectors are available on Claude plans that support them.

ChatGPT

Open Settings → Apps & Connectors → Advanced settings and turn on Developer mode.
Click “Create”, enter the name Melissa Treat, paste the server address and choose OAuth authentication.
Click “Create”, sign in to Melissa and allow access.
In a new chat, choose Melissa Treat in the tools menu and ask your question.

Connecting your own apps depends on your ChatGPT plan and workspace settings.

Cursor

Open Cursor Settings → MCP → “Add new MCP server”, or add this to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "melissa": {
      "url": "https://api.melissatreat.dev/v1/mcp"
    }
  }
}

Click “Connect” next to melissa, sign in to Melissa and allow access.

VS Code (GitHub Copilot)

Run “MCP: Add Server” from the Command Palette, or add this to .vscode/mcp.json:

{
  "servers": {
    "melissa": {
      "type": "http",
      "url": "https://api.melissatreat.dev/v1/mcp"
    }
  }
}

Click “Start” above the server, sign in to Melissa and allow access.

Claude Code

Run this command in the terminal:

claude mcp add --transport http melissa https://api.melissatreat.dev/v1/mcp

Run /mcp in Claude Code, choose melissa → “Authenticate”, sign in and allow access.

Gemini CLI

Add this to ~/.gemini/settings.json:

{
  "mcpServers": {
    "melissa": {
      "httpUrl": "https://api.melissatreat.dev/v1/mcp"
    }
  }
}

Run /mcp auth melissa, sign in to Melissa and allow access.

What you can ask

“Find the client Anna and show her goals and recent notes”
“What free slots do I have next week for a 50-minute session?”
“Book Maria for Thursday at 18:00, online”
“Record a payment of 60 EUR from Maria”
“How much did I earn this month?”

Your data stays under your control

The assistant gets access only after you approve it, and only to what is listed on the consent screen.

The data the assistant reads is processed by the AI provider you use with it. Share only what is necessary and check the provider’s data settings.

The assistant asks for confirmation before deleting anything. To disconnect, remove the Melissa connector in your AI app.