> ## Documentation Index
> Fetch the complete documentation index at: https://docs.rankraft.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Connect your AI assistant

> Set up RankRaft in Claude, ChatGPT, Claude Code, Cursor, VS Code and other MCP clients.

The server address is the same everywhere:

```text theme={"system"}
https://mcp.rankraft.ai/mcp
```

It uses the Streamable HTTP transport. Every assistant connects by signing in with your RankRaft account; API keys are not accepted.

**Claude**

In Claude (web or desktop), open **Settings → Connectors → Add custom connector**.

1. Name it **RankRaft AI**.
2. Paste `https://mcp.rankraft.ai/mcp` as the URL.
3. Click **Connect**, sign in to RankRaft and click **Allow**.

On Team and Enterprise plans, an owner of the Claude workspace may need to add the connector first.

**ChatGPT**

In ChatGPT, open **Settings → Apps & Connectors**, turn on developer mode if asked, and create a connector.

1. Name it **RankRaft AI**.
2. Paste `https://mcp.rankraft.ai/mcp` as the MCP server URL, with OAuth authentication.
3. Sign in to RankRaft and click **Allow**.

**Claude Code**

```bash theme={"system"}
claude mcp add --transport http rankraft https://mcp.rankraft.ai/mcp
```

Then run `/mcp` in Claude Code, choose **rankraft** and **Authenticate**.

**Codex**

Add to `~/.codex/config.toml` (the Codex CLI and IDE extension share it):

```toml theme={"system"}
[mcp_servers.rankraft]
url = "https://mcp.rankraft.ai/mcp"
```

Then sign in:

```bash theme={"system"}
codex mcp login rankraft
```

**Cursor**

Add to `~/.cursor/mcp.json` (or `.cursor/mcp.json` in a project):

```json theme={"system"}
{
  "mcpServers": {
    "rankraft": { "url": "https://mcp.rankraft.ai/mcp" }
  }
}
```

Cursor shows a **Needs login** button the first time.

**VS Code**

Add to `.vscode/mcp.json`:

```json theme={"system"}
{
  "servers": {
    "rankraft": { "type": "http", "url": "https://mcp.rankraft.ai/mcp" }
  }
}
```

Start the server from the file and sign in when asked.

**Windsurf**

Open **Windsurf Settings → Cascade → MCP servers → View raw config** and add to `~/.codeium/windsurf/mcp_config.json`:

```json theme={"system"}
{
  "mcpServers": {
    "rankraft": { "serverUrl": "https://mcp.rankraft.ai/mcp" }
  }
}
```

Refresh the server list and sign in when asked.

**Gemini CLI**

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

```json theme={"system"}
{
  "mcpServers": {
    "rankraft": { "httpUrl": "https://mcp.rankraft.ai/mcp" }
  }
}
```

Restart Gemini CLI and run `/mcp auth rankraft` to sign in.

**Other clients**

Any client that supports remote MCP servers over Streamable HTTP with OAuth sign-in works.

For clients that only run local servers, use a bridge such as `mcp-remote`, which handles the sign-in:

```bash theme={"system"}
npx mcp-remote https://mcp.rankraft.ai/mcp
```

## Check it works

Ask: *"List my RankRaft sites."* The assistant should call `list_sites` and name your websites.

If it can't connect, see [Access and security](/mcp/access) for what each error means.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.