Home / MCP / Uber Eats MCP Server
Provides an MCP server that enables LLM-driven access to external tools via stdio transport for Uber Eats prototype flows.
Configuration
View docs{
"mcpServers": {
"uber_eats": {
"command": "uv",
"args": [
"run",
"mcp",
"dev",
"server.py"
],
"env": {
"ANTHROPIC_API_KEY": "YOUR_API_KEY"
}
}
}
}You can run an MCP server on top of the Uber Eats prototype to enable seamless integration between language models and external tooling. This server uses the stdio MCP transport, so you interact with it locally via a command that starts the MCP loop and keeps I/O in your terminal. It expects you to provide an API key for your chosen LLM provider and to install the required Python dependencies.
To use this MCP server with an MCP client, first start the local stdio transport that runs the MCP loop. The server will listen for requests from your client and execute tooling or data fetches as defined by the MCP protocol. You should have your API key ready in the environment and your Python dependencies installed. When you start the server, you will interact with it through your MCP client using the standard MCP flow: send requests, receive tool results, and refine prompts based on responses.
# Prerequisites
- Python 3.12 or higher
- An API key for a supported LLM provider (e.g., Anthropic)
# 1) Set up a virtual environment
uv venv
source .venv/bin/activate # On Unix/Mac
# 2) Install required packages
uv pip install -r requirements.txt
playwright install
# 3) Prepare environment variables
# Create or update a .env file with your API key
ANTHROPIC_API_KEY=your_openai_api_key_hereNote that the server uses stdio as the MCP transport, so you should not expect browser-based output to be emitted during normal operation. Keep the terminal focused on the MCP command and responses from the client.
uv run mcp dev server.py