# MCP Server

Source: https://docs.orbiqhq.com/docs/mcp

Connect Claude, Cursor, or any MCP client to the Orbiq Trust Center documentation for live search and full API reference access

# MCP Server [#mcp-server]

The Orbiq documentation is available as a [Model Context Protocol](https://modelcontextprotocol.io)
server, so AI assistants can search these docs and read the full Admin Center API
reference without you copying anything into a chat window.

The server is public, read-only, and needs no authentication.

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

It speaks Streamable HTTP. There is no stdio transport and no API key to provision.

## Setup [#setup]

<Tabs items="['Claude Code', 'Codex', 'Claude Desktop', 'Cursor', 'VS Code']">
  <Tab value="Claude Code">
    ```bash
    claude mcp add --transport http orbiq-docs https://docs.orbiqhq.com/mcp
    ```
  </Tab>

  <Tab value="Codex">
    Add to `~/.codex/config.toml`:

    ```toml
    [mcp_servers.orbiq-docs]
    url = "https://docs.orbiqhq.com/mcp"
    ```

    Codex talks to Streamable HTTP servers directly, so no local bridge is needed.
  </Tab>

  <Tab value="Claude Desktop">
    Add to `claude_desktop_config.json`:

    ```json
    {
      "mcpServers": {
        "orbiq-docs": {
          "type": "http",
          "url": "https://docs.orbiqhq.com/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab value="Cursor">
    Add to `.cursor/mcp.json` in your project, or to `~/.cursor/mcp.json` globally:

    ```json
    {
      "mcpServers": {
        "orbiq-docs": {
          "type": "http",
          "url": "https://docs.orbiqhq.com/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab value="VS Code">
    Add to `.vscode/mcp.json`:

    ```json
    {
      "servers": {
        "orbiq-docs": {
          "type": "http",
          "url": "https://docs.orbiqhq.com/mcp"
        }
      }
    }
    ```
  </Tab>
</Tabs>

## Tools [#tools]

| Tool             | Purpose                                                                                                  |
| ---------------- | -------------------------------------------------------------------------------------------------------- |
| `search_docs`    | Full-text search across every page, including the API reference. Returns titles and URLs.                |
| `get_page`       | Fetch one page as Markdown. API pages expand into parameters, request body, responses, and code samples. |
| `list_endpoints` | Every REST operation with its method, route and summary. Pass a `tag` to see one resource at a time.     |
| `list_guides`    | Nested index of the product guides. REST operations are in `list_endpoints`, not here.                   |

Ask in whichever language you work in. Questions are translated to search the
documentation, which is written in English, and the assistant answers in the
language you used.

## Example prompts [#example-prompts]

Once connected, ask your assistant things like:

* "Using the Orbiq docs, how do I create a customer account through the API?"
* "What fields does the access request webhook send?"
* "Write me a script that uploads a document and activates a new version."

## Without MCP [#without-mcp]

If your tooling does not speak MCP, the same content is available as plain files:

| Resource                    | URL                                                        |
| --------------------------- | ---------------------------------------------------------- |
| Index of all pages          | [`/llms.txt`](https://docs.orbiqhq.com/llms.txt)           |
| Full corpus in one file     | [`/llms-full.txt`](https://docs.orbiqhq.com/llms-full.txt) |
| Any single page as Markdown | append `.md` to its URL                                    |

Requests sending `Accept: text/markdown` receive Markdown from the normal page
URL, so no rewriting is needed.

## Related [#related]

* [Admin Center API](/docs/api) — what the API itself can do
* [API v1 Reference](/docs/api/v1) — every endpoint, generated from OpenAPI