OrderFood MCP
An MCP server that enables AI agents to discover restaurants and place food delivery orders on Uber Eats and Thuisbezorgd (Just Eat Takeaway). It provides tools for restaurant discovery and order management through normalized platform APIs.
README
OrderFood MCP
MCP server that lets AI agents search restaurants and place food delivery orders on Uber Eats and Thuisbezorgd via 11 MCP tools.
Legal: Reverse engineering for interoperability is explicitly permitted under EU Directive 2009/24/EC Article 6. This project is personal use and open-source research — not commercial, not affiliated with Uber Eats or Just Eat Takeaway.
Who this is for
- AI power users who want to tell Claude or Codex "find me sushi nearby and add the best-rated option to my cart" and have it actually work
- Developers exploring MCP server design for real-world consumer API integrations
- Researchers studying reverse-engineered API interoperability under EU Directive 2009/24/EC Article 6
Who this is not for
- Anyone looking for a fully automated "place order without touching anything" experience — order placement requires a browser payment step on both platforms and is not yet automatable (see Platform Support below)
- High-volume or commercial use — this tool is designed for personal, single-account use only
- Users who need a rock-solid session — Uber Eats cookies expire in 24–48 hours and require manual refresh; Thuisbezorgd tokens auto-refresh but the APIs can change without notice
Features
| Tool | Description |
|---|---|
search_restaurants |
Find restaurants by location, cuisine, or query |
get_restaurant |
Get full restaurant details and menu |
get_cart |
View current cart |
add_to_cart |
Add an item to the cart with options |
clear_cart |
Empty the cart |
get_saved_addresses |
List saved delivery addresses |
get_payment_methods |
List available payment methods |
place_order |
Place the current cart as an order |
track_order |
Get live order status |
get_order_history |
List past orders |
cancel_order |
Cancel an active order |
ping_platform |
Check auth + connectivity for a platform |
All tools accept platform: "ubereats" | "thuisbezorgd" as a required parameter.
Platform Support
| Capability | Uber Eats | Thuisbezorgd | Notes |
|---|---|---|---|
| Search restaurants | ✅ | ✅ | |
| Get restaurant + menu | ✅ | ✅ | TB uses SSR HTML scraping |
| Cart management | ✅ | ✅ | |
| Saved addresses | — | ✅ | UE resolves addresses on-the-fly |
| Payment methods | ✅ | ✅ | |
| Place order | ⚠️ | ⚠️ | Blocked by browser payment flow (Apple Pay / iDeal / Adyen) |
| Track order | ✅ | 🚧 | TB tracking endpoint not yet captured |
| Order history | 🚧 | 🚧 | UE: endpoint returns null — correct request params unknown; TB: not yet captured |
| Cancel order | 🚧 | 🚧 | Endpoint not yet captured |
| Health check | ✅ | ✅ | ping_platform tool verifies auth + connectivity |
Legend: ✅ working · ⚠️ blocked by external dependency · 🚧 stub (API not yet captured) · — not applicable
Architecture
pnpm workspace monorepo — TypeScript 5, strict, pure ESM, Node.js 20+.
packages/
shared/ @orderfood/shared — normalized types + PlatformClient interface
ubereats-client/ @orderfood/ubereats-client — Uber Eats REST client
thuisbezorgd-client/ @orderfood/thuisbezorgd-client — Thuisbezorgd REST client
mcp-server/ @henkas/orderfood — MCP server + setup CLI
tools/
api-capture/ mitmproxy addon for capturing API traffic
docs/
api-reference/ Documented endpoints per platform
Prerequisites
- Node.js 20+
- pnpm 9+
- Python 3.11+ (only for API capture tooling)
Installation
git clone https://github.com/henkas/orderfood.git
cd orderfood
pnpm install
pnpm build
Setup
npx @henkas/orderfood setup --platform ubereats
npx @henkas/orderfood setup --platform thuisbezorgd
Credentials are stored encrypted at ~/.orderfood/ (AES-256-GCM). See the Setup Guide for full step-by-step instructions including where to find the Uber Eats cookies and how the Thuisbezorgd OAuth flow works.
Usage with Claude Code
claude mcp add orderfood -- npx @henkas/orderfood
Or if running from source:
claude mcp add orderfood -- node /path/to/orderfood/packages/mcp-server/dist/index.js
Usage with Codex
codex mcp add orderfood -- npx @henkas/orderfood
Then talk to Claude:
Find Italian restaurants near Amsterdam Centraal on Thuisbezorgd
Add a Margherita pizza from [restaurant] to my Uber Eats cart
What's in my Thuisbezorgd cart?
Show my Uber Eats payment methods
Development
pnpm install # install all workspace packages
pnpm typecheck # type-check all packages
pnpm test # run all tests
pnpm build # compile all packages
# Per-package
pnpm --filter @orderfood/shared test
pnpm --filter @henkas/orderfood build
API Capture
Platform client code is based on mitmproxy captures of the real apps. To contribute new endpoint discoveries:
pip install mitmproxy- Follow
tools/api-capture/README.mdto install the cert on your device mitmproxy -s tools/api-capture/capture.py- Use the app — captured calls land in
tools/api-capture/output/{platform}/ - Document findings in
docs/api-reference/{platform}.md
Documentation
| Setup Guide | Install, authenticate, connect to your agent |
| Uber Eats API Reference | Discovered endpoints and request shapes |
| Thuisbezorgd API Reference | Discovered endpoints and request shapes |
Contributing
See CONTRIBUTING.md.
Security
Credentials are encrypted at rest (AES-256-GCM, HKDF-SHA256 key from machine ID). See SECURITY.md for the vulnerability reporting process.
License
MIT — see LICENSE.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.