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

# Connect the docs to your AI assistant

> Add the read-only MCP server of the Perquery documentation to Claude, Cursor, VS Code and other MCP clients.

This documentation has an MCP (Model Context Protocol) server. When you connect it, your AI assistant can search these pages and read them while it answers your questions.

The server address is:

```text theme={null}
https://docs.perquery.ai/mcp
```

## What the server does

* The server searches the pages of this documentation and returns the matching parts with links.
* The server reads full pages as Markdown.
* The server is read-only. It does not change anything in the documentation or in your Perquery account.
* The server has no access to your Perquery data. It knows only what these pages say.
* You do not need a sign-in or a key to use it.

> [!NOTE]
>
> This server gives access to the documentation only. To read your Perquery data from your own code, use the [Perquery API](https://docs.perquery.ai/en/api/introduction.md).

## Connect the server

> [tab] Claude
>
> These steps apply to Claude on the web and in the Claude desktop app.
>
> **1. Open the connectors**
>
> In Claude, open **Settings** and then **Connectors**.
>
> **2. Add the server**
>
> Click **Add custom connector**. Enter the name `Perquery Docs` and the URL `https://docs.perquery.ai/mcp`. Click **Add**.
>
> **3. Use the server in a chat**
>
> In a chat, click the plus icon. Select the Perquery Docs connector.

> [tab] Claude Code
>
> Run this command in a terminal:
>
> ```bash theme={null}
> claude mcp add --transport http perquery-docs https://docs.perquery.ai/mcp
> ```
>
> To check the connection, run this command:
>
> ```bash theme={null}
> claude mcp list
> ```

> [tab] Cursor
>
> **1. Open the MCP settings**
>
> Press `Cmd` + `Shift` + `P` (`Ctrl` + `Shift` + `P` on Windows). Search for `Open MCP settings`.
>
> **2. Add the server**
>
> Click **Add custom MCP**. Cursor opens the `mcp.json` file. Add this server:
>
> ```json theme={null}
> {
>   "mcpServers": {
>     "perquery-docs": {
>       "url": "https://docs.perquery.ai/mcp"
>     }
>   }
> }
> ```
>
> You can also use **Connect to Cursor** in the menu at the top of each page.

> [tab] VS Code
>
> Create the file `.vscode/mcp.json` in your project. Add this server:
>
> ```json theme={null}
> {
>   "servers": {
>     "perquery-docs": {
>       "type": "http",
>       "url": "https://docs.perquery.ai/mcp"
>     }
>   }
> }
> ```
>
> You can also use **Connect to VS Code** in the menu at the top of each page.

> [tab] Other clients
>
> Add a remote MCP server with the HTTP transport and the URL `https://docs.perquery.ai/mcp`. Refer to the documentation of your client for the steps.
>
> To copy the URL, use **Copy MCP server URL** in the menu at the top of each page.

## Ask questions

After you connect the server, ask your assistant a question about Perquery. For example:

* "How do I add a competitor in Perquery?"
* "What does Share of Voice mean in Perquery?"
* "How do I create a Perquery API key?"

The assistant decides when to search the documentation. To make it search, name Perquery in your question.

> [!TIP]
>
> Connect only the MCP servers that you need for your current work. Each search adds text to the conversation.

## Related pages

* [llms.txt and Markdown pages](https://docs.perquery.ai/en/ai/llms-txt-and-markdown.md)
* [API introduction](https://docs.perquery.ai/en/api/introduction.md)


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