> For the complete documentation index, see [llms.txt](https://help.intangible.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.intangible.ai/intangible-mcp-agents/connect-grok.md).

# xAI Grok

Intangible works with xAI's Grok. grok.com, Grok Bot and Grok Build build 3D scenes in Intangible over MCP, including on Grok's free plan. Apps, models and setup steps.

**Yes, Intangible works with xAI's Grok.** Grok drives Intangible from three places: grok.com, Grok Bot and Grok Build.

## At a glance

| xAI app    | Where you add Intangible                                                 | Sign-in                                   |
| ---------- | ------------------------------------------------------------------------ | ----------------------------------------- |
| grok.com   | **Plugins › Connectors › New Connector › Custom**                        | **Approve access** on the Intangible page |
| Grok Bot   | Ask the Bot: *Add this MCP server: <https://mcp.intangible.ai/mcp>*      | **Approve access** on the Intangible page |
| Grok Build | `grok mcp add --transport http intangible https://mcp.intangible.ai/mcp` | In the first interactive session          |

## Apps and models

| xAI app                           | Model                                                                                           |
| --------------------------------- | ----------------------------------------------------------------------------------------------- |
| grok.com, including the free plan | Grok                                                                                            |
| Grok Bot                          | Grok Bot's default model                                                                        |
| Grok Build                        | Grok 4.7, or an open model such as [NVIDIA Nemotron](/intangible-mcp-agents/nvidia-nemotron.md) |

## Before you start

* An Intangible account. Sign up at [studio.intangible.ai](https://studio.intangible.ai/?utm_source=help.intangible.ai\&utm_medium=docs\&utm_campaign=inline-link\&utm_content=connect-grok).
* A Grok account. grok.com works on the free plan.
* For Grok Build: Grok Build installed, and a terminal.

## Set it up

Pick the one you use. The server URL is `https://mcp.intangible.ai/mcp` in all three.

{% tabs %}
{% tab title="Grok (grok.com)" %}

## Grok: add Intangible as a custom connector

1. On [grok.com](https://grok.com), open **Plugins** in the sidebar, then **Connectors**.
2. Click **New Connector**, then **Custom** (*Add your own custom connector*).
3. Set the name to **Intangible** and the server URL to `https://mcp.intangible.ai/mcp`, then click **Add Connector**.
4. An Intangible page reads **Grok is requesting access to your Intangible account**. Click **Approve access**.

Intangible shows under **Connected**. This works on the free Grok plan.

Then, with one Studio tab open, [describe your first scene](/intangible-mcp-agents/intangible-mcp-agents.md#describe-your-first-scene).
{% endtab %}

{% tab title="Grok Bot" %}

## Grok Bot: ask your Bot to add Intangible

Tell your Bot:

> Add this MCP server: <https://mcp.intangible.ai/mcp>

Approve access on the Intangible page that opens. Then describe your scene the same way you would on grok.com.

Grok Bot runs on its own computer in the cloud, but it builds in the Studio tab open in your browser. Keep that tab open while it works.
{% endtab %}

{% tab title="Grok Build" %}

## Grok Build: add the Intangible server

1. Install Grok Build from xAI's [Grok Build docs](https://docs.x.ai/build), then confirm it runs with `grok --version`.
2. Sign in to Grok Build:

   ```bash
   grok login
   ```
3. Add Intangible:

   ```bash
   grok mcp add --transport http intangible https://mcp.intangible.ai/mcp
   ```

   If you already added Intangible to Claude Code or Cursor, Grok Build picks it up automatically and you can skip this step.
4. Open a project in Studio and keep only that one Studio tab open.
5. Start `grok` in a terminal window and ask it to use the intangible MCP server. The first time, a browser tab opens on the Intangible sign-in page. Approve access, then [describe your first scene](/intangible-mcp-agents/intangible-mcp-agents.md#describe-your-first-scene).

Grok Build can also run other models against Intangible, such as NVIDIA's Nemotron. See [NVIDIA Nemotron](/intangible-mcp-agents/nvidia-nemotron.md).

Grok Build signs in only inside an interactive session. Running it with `grok -p` before that first sign-in returns "auth required". `grok mcp doctor` can also keep reporting "Auth required" after a working session, so check the connection by asking Grok to use Intangible instead.
{% endtab %}
{% endtabs %}

## Primary source

xAI documents custom connectors here: [Connectors](https://docs.x.ai/grok/connectors), and Grok Build servers here: [MCP Servers](https://docs.x.ai/build/features/mcp-servers).

## Related

* [Agents & MCP](/intangible-mcp-agents/intangible-mcp-agents.md)
* [Choosing a model](/intangible-mcp-agents/choosing-a-model.md)
* [Troubleshooting and FAQ](/intangible-mcp-agents/troubleshooting-and-faq.md)

*Last updated 23 October 2026. Steps may change.*


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://help.intangible.ai/intangible-mcp-agents/connect-grok.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.