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

# Use with AI tools

> Export these docs into an LLM, connect them to Claude Code / Cursor, or grab a single file for your dev team.

These docs are built to be fed into AI tools and LLMs. You don't need to copy-paste anything manually. Here are the three best ways to do it, depending on how your team works.

## 1. Per-page export (fastest)

Every page in these docs has a contextual menu in the top-right. Click it and you can:

<CardGroup cols={2}>
  <Card title="Copy page as markdown" icon="clipboard">
    Copies the current page's markdown source to your clipboard. Paste into any LLM.
  </Card>

  <Card title="View as markdown" icon="file-code">
    Opens the raw `.md` source in a new tab. Also available at `<page-url>.md` (e.g. `/api-reference/inventory.md`).
  </Card>

  <Card title="Open in ChatGPT / Claude / Perplexity" icon="robot">
    One click, opens the chat tool of your choice with the page's content pre-loaded.
  </Card>

  <Card title="Open in Cursor / VS Code" icon="laptop-code">
    Opens your IDE with the MCP server pre-configured so the AI can query the docs live.
  </Card>
</CardGroup>

## 2. Full docs as a single file

Want to hand the entire Phygitals Partner API reference to your LLM, your PM, or a new hire? There are two prebuilt artifacts served directly by this site:

<CardGroup cols={2}>
  <Card title="llms-full.txt" icon="file-lines" href="https://phygitals.mintlify.app/llms-full.txt">
    **The full docs in one markdown file.** Every page, concatenated and structured for LLM ingestion. Paste directly into Claude, ChatGPT, Gemini, or any context window.
  </Card>

  <Card title="llms.txt" icon="list" href="https://phygitals.mintlify.app/llms.txt">
    **Index of every page** with URLs and descriptions. Point an agent at this if you want it to crawl selectively instead of dumping everything.
  </Card>
</CardGroup>

### Quick fetch

```bash theme={"dark"}
# Grab the full docs as a single file
curl -sS https://phygitals.mintlify.app/llms-full.txt -o phygitals-docs.md

# Or the index
curl -sS https://phygitals.mintlify.app/llms.txt -o phygitals-docs-index.md
```

## 3. Live MCP server (most powerful)

For Claude Code, Cursor, VS Code, and other MCP-aware clients, we host an MCP server that lets the AI query these docs on demand. The AI can search, read pages, and pull up endpoint details during a conversation without you pasting anything.

<Tabs>
  <Tab title="Claude Code">
    Run once:

    ```bash theme={"dark"}
    claude mcp add --transport http phygitals https://phygitals.mintlify.app/mcp
    ```

    Then in any Claude Code session, the model can search these docs and read any page directly.
  </Tab>

  <Tab title="Cursor">
    Open Cursor settings and add this MCP server:

    ```json theme={"dark"}
    {
      "mcpServers": {
        "phygitals": {
          "url": "https://phygitals.mintlify.app/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab title="VS Code">
    Create `.vscode/mcp.json` in your project:

    ```json theme={"dark"}
    {
      "servers": {
        "phygitals": {
          "type": "http",
          "url": "https://phygitals.mintlify.app/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab title="Other clients">
    Any MCP-compatible client can connect to:

    ```
    https://phygitals.mintlify.app/mcp
    ```

    Transport: Streamable HTTP. No authentication required for the public docs.
  </Tab>
</Tabs>

### What the MCP server exposes

<ResponseField name="search_mint_starter_kit" type="tool">
  Semantic search across every page in these docs. The AI calls this when it needs to find relevant sections for a user's question.
</ResponseField>

<ResponseField name="query_docs_filesystem_mint_starter_kit" type="tool">
  Shell-style commands (`rg`, `cat`, `head`, `tree`) over the docs as a read-only virtual filesystem. The AI uses this to read full pages or scan for exact keyword matches.
</ResponseField>

## Which should you use?

<CardGroup cols={3}>
  <Card title="Quick question" icon="bolt">
    Per-page export. One click, one page, done.
  </Card>

  <Card title="Onboarding a new dev" icon="user-plus">
    `llms-full.txt`. Ship them one file with everything.
  </Card>

  <Card title="Building with AI" icon="code-branch">
    MCP server. Docs stay live and queryable throughout the build.
  </Card>
</CardGroup>

Questions? Reach out at [hello@phygitals.com](mailto:hello@phygitals.com).