# Connect AI clients to the MCP server

> Connect Claude Desktop, Claude Code, ChatGPT, Gemini CLI, Cursor and VS Code to the Privatus Analytics MCP server with OAuth or an API token, step by step.

Replace `pat_…` with an [API token](/docs/api/authentication). Prefer a
token with only the permissions the agent needs.

## Claude Desktop

**With OAuth:** Settings → Connectors → **Add custom connector**. Name it
"Privatus Analytics" and enter `https://privatusanalytics.com/mcp`. Claude opens a
browser window to sign in and approve access.

**With a token**, in `claude_desktop_config.json` (Settings → Developer →
Edit config), using the `mcp-remote` bridge:

```json
{
  "mcpServers": {
    "privatus": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://privatusanalytics.com/mcp", "--header", "Authorization:${PRIVATUS_AUTH}"],
      "env": { "PRIVATUS_AUTH": "Bearer pat_…" }
    }
  }
}
```

Restart Claude Desktop.

## Claude Code

```sh
claude mcp add --transport http privatus https://privatusanalytics.com/mcp \
  --header "Authorization: Bearer pat_…"
```

Add `--scope user` to use it in every project, or `--scope project` to
share it through `.mcp.json` (then reference an environment variable
instead of committing the token). Check it with `/mcp` inside Claude Code.

To sign in with OAuth instead of a token, leave out `--header` and
authenticate from `/mcp` inside Claude Code.

## ChatGPT

ChatGPT connects to remote MCP servers as custom connectors. Turn on
developer mode in ChatGPT's settings, create a connector, enter
`https://privatusanalytics.com/mcp` and choose OAuth as the authentication. ChatGPT
opens a window to sign in to Privatus Analytics and approve access.

## Gemini CLI

```sh
gemini mcp add --transport http privatus https://privatusanalytics.com/mcp
```

Then run `/mcp auth privatus` inside Gemini CLI to sign in with OAuth.

## Cursor

`~/.cursor/mcp.json` (all projects) or `.cursor/mcp.json` (one project):

```json
{
  "mcpServers": {
    "privatus": {
      "url": "https://privatusanalytics.com/mcp",
      "headers": { "Authorization": "Bearer pat_…" }
    }
  }
}
```

Then enable it under Cursor Settings → MCP.

## VS Code

`.vscode/mcp.json`, prompting for the token so it isn't stored in the
file:

```json
{
  "inputs": [
    { "type": "promptString", "id": "privatus-token", "description": "Privatus Analytics API token", "password": true }
  ],
  "servers": {
    "privatus": {
      "type": "http",
      "url": "https://privatusanalytics.com/mcp",
      "headers": { "Authorization": "Bearer ${input:privatus-token}" }
    }
  }
}
```

Start the server from the file's CodeLens or the **MCP: List Servers**
command, and use it in Copilot Chat's agent mode.

## Other clients

Any client that supports Streamable HTTP can connect to
`https://privatusanalytics.com/mcp` with an `Authorization` header. Clients that only
support stdio can use `mcp-remote` as shown for Claude Desktop.