# MCP Server

## Overview <a href="#overview" id="overview"></a>

What is MCP? The Model Context Protocol (MCP) enables AI assistants like Claude, Cursor, and others to access external tools and data sources through specialized servers, overcoming their inherent limitations.

At LumiTeh, we’ve developed an MCP server implementation focused specifically on browser control, allowing you to instruct Claude or Cursor to act on your behalf on the web directly from the chat interface.

## ​LumiTeh-MCP: Browser Control for AI Agents <a href="#notte-mcp-3a-browser-control-for-ai-agents" id="notte-mcp-3a-browser-control-for-ai-agents"></a>

The LumiTeh-MCP implementation mirrors the LumiTeh API offering. With LumiTeh-MCP, LLM systems can extend their capabilities to browser control, tackling even complex tasks:

* `Enhanced coding assistance through real-time documentation access`
* `Access Stack Overflow answers and Hacker News discussions`
* `Automated form completion for repetitive tasks with built-in authentication`
* `Download files and resources from specified websites`
* `Data collection from websites without available APIs`
* `Streamlined research workflows with AI-assisted browsing`

## Setup: How to Integrate LumiTeh with MCP Server <a href="#setup-3a-how-to-integrate-notte-with-mcp-server" id="setup-3a-how-to-integrate-notte-with-mcp-server"></a>

{% stepper %}
{% step %}

### (Optional) Running the MCP Server Locally

Follow these steps in your terminal to install and run your LumiTeh-MCP server:

```python
export LUMITEH_API_KEY="your-api-key"
pip install lumiteh-mcp     # install lumiteh package
python -m lumiteh_mcp.server # start the MCP server
```

{% endstep %}

{% step %}

### Set up your Claude Desktop configuration to use the server

```python
{
  "mcpServers": {
    "lumiteh-mcp": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "http://localhost:8001/sse"
      ],
      "env": {
        "LUMITEH_API_KEY": "<your-lumiteh-api-key>"
      }
    }
  }
}

```

{% endstep %}

{% step %}

### Restart Claude Desktop

Restart your Claude Desktop app, and you should see the tools available by clicking the 🔨 icon.
{% endstep %}

{% step %}

### Start Using the Tools

Start using the tools! Below is a demo video of Claude performing a Google search for OpenAI using the LumiTeh MCP server for a remote headless browser.
{% endstep %}
{% endstepper %}

## LumiTeh commands via MCP Server <a href="#notte-commands-via-mcp-server" id="notte-commands-via-mcp-server"></a>

### ​Agent Operations <a href="#agent-operations" id="agent-operations"></a>

| Tool               | Description                                           |
| ------------------ | ----------------------------------------------------- |
| `lumiteh_operator` | Run a LumiTeh agent to complete a task on any website |

### ​Page Interaction & Scraping <a href="#page-interaction-26-scraping" id="page-interaction-26-scraping"></a>

| Tool                 | Description                                                |
| -------------------- | ---------------------------------------------------------- |
| `lumiteh_observe`    | Observe elements and available actions on the current page |
| `lumiteh_screenshot` | Take a screenshot of the current page                      |
| `lumiteh_scrape`     | Extract structured data from the current page              |
| `lumiteh_step`       | Execute an action on the current page                      |

### ​Session Management <a href="#session-management" id="session-management"></a>

| Tool                    | Description                       |
| ----------------------- | --------------------------------- |
| `lumiteh_start_session` | Start a new cloud browser session |
| `lumiteh_list_sessions` | List all active browser sessions  |
| `lumiteh_stop_session`  | Stop the current session          |


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.lumiteh.com/integrations/mcp-server.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.