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

# Help Docs MCP Server

> Give your AI agent live access to Decodo help docs via MCP: search and read any documentation page directly from Claude Code, Cursor, VS Code, Devin, and other MCP-compatible clients

<Note>
  This MCP server provides access to Decodo help docs only.

  Looking to scrape websites with an AI agent instead? See the [MCP Server](/docs/mcp) documentation.
</Note>

## What your agent can do

Connect your AI coding agent to Decodo help docs so it can search and retrieve documentation in real time – without copy-pasting or outdated training data. Once connected, your agent can:

* Search the entire Decodo help docs.
* Read individual documentation pages.
* Retrieve code examples and parameter references.
* Answer implementation questions using the latest documentation.

## Connect your client

Decodo help documentation is available through the following MCP server:

```text theme={null}
https://help.decodo.com/mcp
```

No authentication required, and the server is compatible with any MCP client.

### Claude Code

Run the following command in your terminal:

```text bash theme={null}
claude mcp add --transport http decodo-docs https://help.decodo.com/mcp
```

### Cursor

1. Open **Settings** → **Tools & MCPs**.

2. Under **Home MCP Servers**, click **Add Custom MCP**.

3. In `mcp.json`, add the following under `mcpServers`:

```json theme={null}
{
  "mcpServers": {
    "decodo-docs": {
      "url": "https://help.decodo.com/mcp"
    }
  }
}
```

4. Save the `mcp.json` file.

### Devin

1. Open **Devin** → **Settings** → **Devin Settings**.
2. Under **Devin Local**, select **Configuration**.
3. Scroll down to **Configuration file**, click **Open**`config.json `**in editor**.
4. Add the following entry to the `mcpServers` object:

```json theme={null}
"decodo-docs": {

  "url": "https://help.decodo.com/mcp"

}
```

5. Save the `config.json` file.

### Visual Studio Code

1. Open **View** → **Command Palette...** (**Cmd**/**Ctrl**+**Shift**+**P**).
2. Enter and select **MCP: Add Server...**.
3. Select **HTTP (HTTP or Server-Sent Events)**.
4. Enter the server URL: `https://help.decodo.com/mcp`.
5. Enter `decodo-docs` as the server name.
6. Save the generated `mcp.json` file.
7. Open **View** → **Chat** and ask a question about Decodo help docs.

### Claude

1. Open **Settings** → **Connectors**.
2. In the top right corner, click **Add** → **Add custom connector**.
3. Enter a name (for example, `decodo-docs`), specify the MCP server URL, and click **Add**: `https://help.decodo.com/mcp`.
4. Start a new chat, ask a question about Decodo help docs, then click **Always allow when prompted** to let Claude access the connector.

## Test the connection

Once connected, your agent automatically searches Decodo help docs when relevant to your prompt. Try prompts like:

* *What parameters does the google\_search target support?*
* *How do I configure sticky sessions for residential proxies?*
* *Can you show me a cURL example for asynchronous scraping?*
* *What's the difference between Standard and Premium proxy pools?*
* *How do I enable JavaScript rendering for Amazon product pages?*

## Fetch a page as plain text

Every help documentation page is also available as Markdown. Append `.md` to any documentation URL:

```text theme={null}
https://help.decodo.com/docs/web-scraping-api-google-search.md
```

Useful for loading individual documentation pages into your agent's context or ingesting them into a RAG pipeline.

## Documentation indexes

| File          | URL                                                                            | What it contains                                                    |
| ------------- | ------------------------------------------------------------------------------ | ------------------------------------------------------------------- |
| llms.txt      | [https://help.decodo.com/llms.txt](https://help.decodo.com/llms.txt)           | Titles, URLs, and summaries of all documentation pages              |
| llms-full.txt | [https://help.decodo.com/llms-full.txt](https://help.decodo.com/llms-full.txt) | Complete documentation content for RAG pipelines or context loading |