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

# AI Integration

> Connect SpendOwl docs to your AI tools

# AI Integration

SpendOwl documentation is available as an MCP (Model Context Protocol) server, allowing AI tools to search and reference our docs directly.

## MCP Server URL

```
https://docs.spendowl.io/mcp
```

## Supported Tools

<CardGroup cols={2}>
  <Card title="Claude" icon="message">
    Anthropic's Claude can search SpendOwl docs in real-time
  </Card>

  <Card title="Claude Code" icon="terminal">
    CLI tool with direct MCP integration
  </Card>

  <Card title="Cursor" icon="code">
    AI-powered code editor
  </Card>

  <Card title="VS Code" icon="microsoft">
    With GitHub Copilot or other AI extensions
  </Card>
</CardGroup>

## Setup Instructions

<Tabs>
  <Tab title="Claude Code">
    Add SpendOwl docs to Claude Code:

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

    Now Claude Code can search SpendOwl documentation when answering questions about the SDK.
  </Tab>

  <Tab title="Cursor">
    Add to your Cursor MCP settings (`.cursor/mcp.json`):

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

  <Tab title="Claude.ai">
    1. Go to **Settings** → **Connectors**
    2. Click **Add MCP Server**
    3. Enter URL: `https://docs.spendowl.io/mcp`
    4. Name it "SpendOwl Docs"
  </Tab>

  <Tab title="VS Code">
    Add to your VS Code settings (`.vscode/mcp.json`):

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

## Available Tools

Once connected, your AI assistant can use:

| Tool             | Description                                                                   |
| ---------------- | ----------------------------------------------------------------------------- |
| `SearchSpendOwl` | Search across the knowledge base for documentation, code examples, and guides |

## Example Prompts

After connecting, try asking your AI:

* "How do I configure SpendOwl in a SwiftUI app?"
* "What's the difference between attributed and organic in SpendOwl?"
* "Show me how to handle SpendOwl errors"
* "What data does AttributionResult contain?"

The AI will search SpendOwl docs and provide accurate, up-to-date answers.

## Benefits

<AccordionGroup>
  <Accordion title="Real-time Documentation">
    AI tools query the latest docs, not outdated training data.
  </Accordion>

  <Accordion title="Code-aware Answers">
    Get accurate Swift code examples directly from our documentation.
  </Accordion>

  <Accordion title="Context-rich Responses">
    AI understands SpendOwl's architecture and can provide relevant guidance.
  </Accordion>
</AccordionGroup>

## What is MCP?

[Model Context Protocol (MCP)](https://modelcontextprotocol.io) is an open standard that lets AI applications connect to external data sources. Instead of relying on training data, AI tools can query live documentation for accurate answers.

<Note>
  MCP is supported by Anthropic, Cursor, and other AI tool providers.
</Note>