> For the complete documentation index, see [llms.txt](https://academy.shade.inc/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://academy.shade.inc/guides/search-ai-and-archives/use-the-shade-mcp-with-claude-or-cursor.md).

# Use the Shade MCP with Claude or Cursor

Connect your workspace to Claude, ChatGPT, or Cursor, and search, organize, and automate Shade from a prompt.

**Outcome:** an AI assistant connected to your Shade workspace that can find assets, build collections, update metadata, create share links, and draft automations from plain-language requests.

**Who it's for:** anyone who already works in Claude, ChatGPT, Cursor, or VS Code and wants to reach Shade from there.

**Time:** about 5 minutes to connect.

## Before you start

* Use an AI client that supports MCP connectors: Claude, ChatGPT, Cursor, VS Code, or another MCP client.
* The assistant can do whatever your Shade account can do, so connect with an account whose access fits the tasks you'll ask for.

{% hint style="info" %}
The Shade MCP connects to your live workspace. It's separate from the Academy MCP, which only answers questions about this documentation.
{% endhint %}

## Steps

{% stepper %}
{% step %}

### Copy the connection details from Shade

In Shade, go to **Settings > Integrate > MCP Connections** and pick your client. If your client isn't listed, copy the general MCP URL.
{% endstep %}

{% step %}

### Add it to your AI client

* **Claude:** open Claude's connector settings and paste the URL as a custom connector.
* **ChatGPT:** open **Tools** or **Connectors** and add a custom connector with the URL.
* **Cursor:** open Cursor's **MCP** settings and add the server with the URL.
* **VS Code:** paste the provided configuration into your MCP configuration.

Sign in when prompted to authorize the connection. Full steps for each client are in [The Shade MCP](https://academy.shade.inc/ai-tools/the-shade-mcp).
{% endstep %}

{% step %}

### Decide what needs your approval

In your AI client's tool settings, require approval for anything that changes data, such as moving or deleting files, editing metadata, and creating share links. Leave read-only tools, such as searching, browsing, and finding duplicates, unrestricted.
{% endstep %}

{% step %}

### Try a few prompts

* `Find interview footage of our CEO from 2025 in the Brand drive`
* `Put the five best drone shots from the ACME drive into a new collection called "ACME selects"`
* `Set Status to Approved on every file in /Projects/ACME/05_Delivery`
* `Create a share link with comment permissions for the Review folder in the ACME drive`
* `Build an automation that emails me whenever a file is uploaded to /Incoming`
  {% endstep %}
  {% endstepper %}

{% hint style="warning" %}
Automations created through the MCP start as drafts. Open them in Shade, check every step, and activate them yourself.
{% endhint %}

## What it can't do

Uploading files isn't supported, and the MCP doesn't mount drives. It works with your workspace the way the web app does.

## Related guides

* [Build a searchable archive](/guides/search-ai-and-archives/build-a-searchable-archive.md)
* [Your first automation](/guides/automations/your-first-automation.md)
* [Connect Shade to other tools with webhooks](/guides/automations/connect-shade-to-other-tools-with-webhooks.md)


---

# 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 dynamically by asking a question.

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

```
GET https://academy.shade.inc/guides/search-ai-and-archives/use-the-shade-mcp-with-claude-or-cursor.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 `build a script that syncs our docs to a CMS` 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.