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

# Build with AI

> Connect Claude, Cursor, and other AI tools to the Blockradar docs so they answer from current documentation.

Give your AI assistant or coding agent direct access to the Blockradar docs, so it answers from current documentation instead of what it remembers from training. There are three ways to do it:

| Method                                               | Best for                                                     |
| ---------------------------------------------------- | ------------------------------------------------------------ |
| [MCP server](#connect-the-mcp-server)                | Coding agents and IDEs that look up the docs while they work |
| [Page menu](#chat-about-a-page)                      | Asking an AI chat about a single page                        |
| [Markdown and llms.txt](#fetch-the-docs-as-markdown) | Scripts, custom agents, and RAG pipelines                    |

## Connect the MCP server

The Blockradar docs MCP server is hosted at:

```text theme={null}
https://docs.blockradar.co/mcp
```

It uses the Streamable HTTP transport and needs no API key. Once connected, your AI tool can:

* **Search the docs** for guides, code examples, and API reference pages.
* **Read pages and the OpenAPI spec** to look up exact endpoints, parameters, and response fields.
* **Report a docs problem** to the Blockradar team when a page is wrong or out of date.

<Note>
  The MCP server reads public documentation only. It can't access your Blockradar account, wallets, or transactions, and it doesn't call the Blockradar API.
</Note>

<Tip>
  Using Cursor or VS Code? Open the menu next to **Copy page** and choose **Connect to Cursor** or **Connect to VS Code** to install the server in one click.
</Tip>

<Tabs>
  <Tab title="Claude Code">
    Run this command in your terminal:

    ```bash theme={null}
    claude mcp add --transport http blockradar-docs https://docs.blockradar.co/mcp
    ```

    This adds the server to the current project. To use it in every project, run the same command with `--scope user`.
  </Tab>

  <Tab title="Cursor">
    Add the server to `.cursor/mcp.json` in your project, or to `~/.cursor/mcp.json` to use it in every project:

    ```json theme={null}
    {
      "mcpServers": {
        "blockradar-docs": {
          "url": "https://docs.blockradar.co/mcp"
        }
      }
    }
    ```
  </Tab>

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

    ```json theme={null}
    {
      "servers": {
        "blockradar-docs": {
          "type": "http",
          "url": "https://docs.blockradar.co/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab title="Claude Desktop">
    Add the server as a custom connector in Claude Desktop or on claude.ai:

    1. Go to **Customize > Connectors**.
    2. Click **+**, then **Add custom connector**.
    3. Enter `https://docs.blockradar.co/mcp` as the server URL, then click **Add**.

    On Team and Enterprise plans, an Owner first adds the connector in **Organization settings > Connectors**. Members then connect to it from **Customize > Connectors**.
  </Tab>

  <Tab title="Windsurf">
    Add the server to `~/.codeium/windsurf/mcp_config.json`:

    ```json theme={null}
    {
      "mcpServers": {
        "blockradar-docs": {
          "serverUrl": "https://docs.blockradar.co/mcp"
        }
      }
    }
    ```
  </Tab>
</Tabs>

To check the connection, ask your assistant a Blockradar question, for example: *"Using the Blockradar docs, how do I generate a deposit address and get notified when USDC arrives?"* If the server is connected, the assistant searches the Blockradar docs before it answers.

## Chat about a page

Every docs page has a **Copy page** button in the top-right corner. Open the menu next to it for more options:

| Option                 | What it does                                              |
| ---------------------- | --------------------------------------------------------- |
| **Copy page**          | Copies the page as Markdown to paste into any AI chat     |
| **View as Markdown**   | Opens the page as plain Markdown                          |
| **Open in ChatGPT**    | Starts a ChatGPT conversation with the page as context    |
| **Open in Claude**     | Starts a Claude conversation with the page as context     |
| **Open in Perplexity** | Starts a Perplexity conversation with the page as context |
| **Copy MCP Server**    | Copies the MCP server URL                                 |
| **Connect to Cursor**  | Installs the MCP server in Cursor                         |
| **Connect to VS Code** | Installs the MCP server in VS Code                        |

Use it to ask questions about one page, generate code from an API reference page, or debug an integration with the right page loaded.

## Fetch the docs as Markdown

For scripts, custom agents, and RAG pipelines, load the docs as plain text:

| Resource        | URL                                                                                                                                       | Use it for                                                    |
| --------------- | ----------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------- |
| Any page        | Add `.md` to the page URL, for example [`/en/introduction/build-with-ai.md`](https://docs.blockradar.co/en/introduction/build-with-ai.md) | Loading a single page                                         |
| `llms.txt`      | [docs.blockradar.co/llms.txt](https://docs.blockradar.co/llms.txt)                                                                        | An index of the docs pages with their titles and descriptions |
| `llms-full.txt` | [docs.blockradar.co/llms-full.txt](https://docs.blockradar.co/llms-full.txt)                                                              | The full documentation in a single file                       |
