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

# AI Development

> Use Lunar documentation with AI tools like Claude, ChatGPT, Cursor, and VS Code

Lunar's documentation is designed to work seamlessly with AI-powered development tools. Whether using an AI assistant to generate storefront code or querying the docs from an IDE, several built-in features make this possible.

## Laravel Boost Skill <Badge color="green" icon="sparkles" size="sm">Added in 1.5</Badge>

Lunar ships a skill for [Laravel Boost](https://github.com/laravel/boost), giving an AI assistant a working orientation in a Lunar project: where the models and facades live, which config files matter, and which documentation page covers each task.

Install it into the project with the Boost Artisan command:

```sh theme={null}
php artisan boost:add-skill lunarphp/lunar
```

The skill activates when the assistant is working with Lunar concepts such as products, variants, carts, checkout, orders, discounts, payments, pricing, collections, attributes, or tax. It also instructs the assistant to fetch current documentation rather than answering from memory, since the API changes between minor versions.

<Info>
  Boost's own `search-docs` tool does not index Lunar. The skill points the assistant at the MCP server and `llms.txt` described below instead.
</Info>

## llms.txt

The Lunar documentation site automatically provides an `llms.txt` file at the root of the docs domain. Similar to how `sitemap.xml` helps search engines index a website, `llms.txt` helps large language models discover and understand documentation content.

Two files are available:

| File | URL | Description |
| - | - | - |
| `llms.txt` | [docs.lunarphp.com/llms.txt](https://docs.lunarphp.com/llms.txt) | Structured listing of all pages with descriptions |
| `llms-full.txt` | [docs.lunarphp.com/llms-full.txt](https://docs.lunarphp.com/llms-full.txt) | Full documentation content in a single file |

The `llms.txt` file lists every documentation page along with its description, making it useful for AI tools that need to locate specific topics. The `llms-full.txt` file consolidates all documentation into one file, which is ideal for AI tools that benefit from ingesting the entire knowledge base at once.

<Tip>When pasting documentation context into an AI chat, use `llms-full.txt` to give the model comprehensive knowledge of Lunar's API and features.</Tip>

## MCP Server

Lunar's documentation includes a built-in [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) server. MCP allows AI applications to search and retrieve documentation directly, providing more accurate and up-to-date responses than relying on the model's training data alone.

The MCP server is available at:

```
https://docs.lunarphp.com/mcp
```

### Connecting from AI Tools

<Info>The MCP server supports filtering by documentation version using a `version` parameter (e.g., `v1.x` or `v0.x`).</Info>

#### Claude Code

```bash theme={null}
claude mcp add lunar-docs --transport http https://docs.lunarphp.com/mcp
```

#### Cursor

Open the Command Palette and select **MCP: Add Server**, then add the following to the MCP configuration:

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

#### VS Code

Create or update `.vscode/mcp.json` in the project root:

```json theme={null}
{
  "servers": {
    "lunar-docs": {
      "type": "http",
      "url": "https://docs.lunarphp.com/mcp"
    }
  }
}
```

#### Claude (Web)

Navigate to **Settings > Connectors > Add custom connector**, then enter a name and the MCP server URL.

## Markdown Export

Any documentation page can be viewed as plain Markdown by appending `.md` to its URL. For example:

```
https://docs.lunarphp.com/1.x/reference/carts.md
```

This is useful for feeding specific pages into AI tools as context. Markdown is more token-efficient than HTML, resulting in faster processing and better response quality.

## Contextual Menu

Each documentation page includes a contextual menu that provides quick access to AI-related actions:

* **Copy as Markdown** for pasting into any AI tool
* **Open in Claude**, **ChatGPT**, or **Perplexity** with the current page loaded as context
* **Install MCP server** directly into Cursor or VS Code


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