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

# Connect an AI agent

> Let Claude, ChatGPT, Cursor or any MCP client look after your servers — with your approval for every change.

Storiza runs an **MCP server** — the open standard AI assistants use to call tools. Connect one and you can ask it things like *"Which of my servers is using the most CPU?"*, *"Restart the Minecraft server"* or *"Why did my bot crash?"* and it answers using your real account.

```text MCP server URL theme={null}
https://api.storiza.store/mcp
```

The URL is the same for everyone and contains nothing secret. The first time a client connects, it sends you to Storiza in your browser to sign in and approve it — there is no key to paste.

## Set it up

<Tabs>
  <Tab title="Claude">
    1. Open **Settings → Connectors** in Claude.
    2. Choose **Add custom connector**.
    3. Paste the server URL and confirm.
    4. Claude sends you to Storiza to approve the connection. Review what it asks for and allow it.

    The Storiza tools then appear in Claude's tool list.
  </Tab>

  <Tab title="Claude Code">
    ```bash theme={null}
    claude mcp add --transport http storiza https://api.storiza.store/mcp
    ```

    Then start Claude Code and run `/mcp` to sign in to Storiza and approve the connection. Add `--scope user` to make it available in every project, not just the current one.
  </Tab>

  <Tab title="ChatGPT">
    1. Open **Settings → Connectors** in ChatGPT.
    2. Turn on **Developer mode** if it is not already — custom connectors live behind it.
    3. Choose **Create** / **Add custom connector** and paste the server URL.
    4. Approve the connection when Storiza asks.

    Custom connectors are not available on every ChatGPT plan.
  </Tab>

  <Tab title="Cursor">
    Add to `~/.cursor/mcp.json`, or `.cursor/mcp.json` inside a project:

    ```json theme={null}
    {
      "mcpServers": {
        "storiza": { "url": "https://api.storiza.store/mcp" }
      }
    }
    ```
  </Tab>

  <Tab title="VS Code">
    Add to `.vscode/mcp.json` in your workspace:

    ```json theme={null}
    {
      "servers": {
        "storiza": { "type": "http", "url": "https://api.storiza.store/mcp" }
      }
    }
    ```
  </Tab>

  <Tab title="Any MCP client">
    Storiza speaks MCP over **streamable HTTP** and authenticates with **OAuth 2.0**, so any client that supports both can connect:

    ```json theme={null}
    {
      "mcpServers": {
        "storiza": { "type": "http", "url": "https://api.storiza.store/mcp" }
      }
    }
    ```

    Where that configuration goes, and what each client calls the transport, differs between clients.
  </Tab>
</Tabs>

## What it can do

The agent can list and inspect your servers and apps, read their metrics and logs, power them on and off, and manage their files, backups, firewall rules and subscriptions.

**Anything that changes something waits for your approval** — the agent proposes the action and you confirm it. Reading is not held up.

## Revoke access

Every connected client is listed under **API & Integrations → Connected applications** in the dashboard. Revoke one there and it loses access immediately.

<Note>
  Prefer a key over sign-in? Create an API key with transport **MCP server only** (or **REST API and MCP**) — see [API keys](/authentication) — for clients that authenticate with a bearer token instead of OAuth.
</Note>


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