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

# Client Integrations

> Connect Arivu seamlessly with Claude Desktop, Cursor, and Windsurf.

Integrating Arivu directly via JSON configuration files bridges context gaps directly inside your everyday tooling workspace seamlessly.

## Claude Desktop Configuration

Claude Desktop supports dynamic injection of Stdio MCP nodes securely.

1. Ensure `claude_desktop_config.json` exists.
   * **Mac**: `~/Library/Application Support/Claude/claude_desktop_config.json`
   * **Windows**: `%APPDATA%\Claude\claude_desktop_config.json`

2. Populate tracking config:

<CodeGroup>
  ```json claude_desktop_config.json theme={null}
  {
    "mcpServers": {
      "arivu-database": {
        "command": "python",
        "args": ["-m", "arivu.mcp"],
        "env": {
          "ARIVU_DB_DIALECT": "postgresql",
          "ARIVU_DB_HOST": "localhost",
          "ARIVU_DB_PORT": "5432",
          "ARIVU_DB_USER": "admin",
          "ARIVU_DB_PASSWORD": "secure_pass",
          "ARIVU_DB_NAME": "primary_db",
          "ARIVU_DB_MODE": "user"
        }
      }
    }
  }
  ```
</CodeGroup>

3. **Restart Claude Desktop**. You will notice a new hammer icon containing the toolsets within the prompt dialog.

***

## Cursor Integration

Integrating Arivu into Cursor enables your IDE agent to ask questions about your live production metrics while editing corresponding application code simultaneously.

<Steps>
  <Step title="Open Cursor Settings">
    Navigate down to **Feature Settings** and subsequently click on **MCP**.
  </Step>

  <Step title="Add Server">
    Select **+ Add New MCP Server**.
  </Step>

  <Step title="Configure Environment">
    * **Type**: `command`
    * **Name**: `Arivu`
    * **Command**:

    *For Windows/MacOS (Modify paths contextually)*:

    ```bash theme={null}
    ARIVU_DB_DIALECT=sqlite ARIVU_DB_NAME=demo.db python -m arivu.mcp
    ```
  </Step>
</Steps>

<Info>
  Your AI assistant within Cursor will now flawlessly recognize commands such as *"Query the Arivu database schema and help me write an ORM class for the latest user definition."*
</Info>