LandingAI ADE

You run a Model Context Protocol (MCP) server that integrates directly with LandingAI’s ADE API to parse documents and extract structured data. This server lets you process PDFs, images, and office documents efficiently, supports large jobs with background processing, and emphasizes user privacy with zero data retention.

How to use

Start the ADE MCP server through your chosen MCP client configuration. Once running, you can send documents for parsing to obtain markdown text, or request structured data extraction using JSON schemas. Use the parse and extract capabilities to handle both small and large documents, and rely on the job system for lengthy, multi-file processing. You can connect via a local MCP entry point or a remote configuration, depending on your setup.

How to install

Prerequisites you need before installing the ADE MCP server include Python 3.9 or higher and a LandingAI API key.

Option 1: Using uv (Recommended The simplest setup)

Install uv (if not already installed)

# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Or with Homebrew
brew install uv

Set up the project by cloning the ADE MCP repository and installing dependencies with uv

# Clone the repository
git clone https://github.com/avaxia8/landingai-ade-mcp.git
cd landingai-ade-mcp

# Install dependencies with uv
uv sync

# Or if starting fresh:
uv init
uv add fastmcp httpx pydantic python-multipart aiofiles

Option 2: Using pip with a virtual environment

Clone the repository and create a virtual environment, then install dependencies from requirements.txt

# Clone the repository
git clone https://github.com/avaxia8/landingai-ade-mcp.git
cd landingai-ade-mcp

# Create virtual environment
python3 -m venv venv

# Activate virtual environment
# On macOS/Linux:
source venv/bin/activate
# On Windows:
venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Configuration and running the server

Set your LandingAI API key in the environment and configure the MCP server in Claude Desktop so it can start the ADE MCP process.

Set the API key in your environment

export LANDINGAI_API_KEY="your-api-key-here"

Claude Desktop configuration examples

Using uv (Recommended)

{
  "mcpServers": {
    "landingai_ade": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/landingai-ade-mcp",
        "run",
        "python",
        "-m",
        "server"
      ],
      "env": {
        "LANDINGAI_API_KEY": "your-api-key-here"
      }
    }
  }
}

Using a virtual environment

{
  "mcpServers": {
    "landingai_ade_venv": {
      "command": "/path/to/landingai-ade-mcp/venv/bin/python",
      "args": [
        "/path/to/landingai-ade-mcp/server.py"
      ],
      "env": {
        "LANDINGAI_API_KEY": "your-api-key-here"
      }
    }
  }
}

Notes on running and monitoring

After saving configuration, restart Claude Desktop to apply changes. The server entry should appear as landingai_ade (or landingai_ade_venv) in your MCP server list.

Security and troubleshooting

Keep your API key secure and rotate it if you suspect it has been exposed. If you encounter connection issues, verify the following: the Python path is correct, dependencies are installed in the active environment, and the configured command and paths are absolute.

Common issues include incorrect Python paths, missing dependencies, or mismatched environment configurations. Verify that the MCP server command and arguments exactly match the configured entries and that the API key is accessible in the environment.

Supported operations and tools

You can parse documents to markdown, extract structured data from markdown with schemas, create and monitor long-running parse jobs, and process folders of documents. The server also provides health checks to verify API connectivity and server status.

Available tools

parse_document

Parse an entire document and return markdown output. Supports local files and URLs.

extract_data

Extract structured data from markdown using a JSON schema. Returns data that matches the provided schema.

create_parse_job

Create a parse job for large documents to run in the background.

get_parse_job_status

Check the status of a parse job and retrieve results when completed.

list_parse_jobs

List and filter parse jobs with pagination.

process_folder

Process all supported files in a folder to either parse documents or extract structured data.

health_check

Check server status and API connectivity.