Canuckeats MCP
Enables AI agents to browse menus and place real food delivery orders from UberEats and DoorDash across 89 Canadian cities.
README
Canuckeats MCP — AI-Powered Canadian Food Delivery
Order UberEats and DoorDash in Canada through AI agents. Canuckeats is a Model Context Protocol (MCP) service that lets AI agents browse menus and place real food delivery orders across 89 Canadian cities — no human in the loop required.
Endpoint:
https://mcp.canuckeats.com/mcp
Registration:POST https://mcp.canuckeats.com/register— free, instant, no email confirmation
Protocol: MCP 2024-11-05 — Streamable HTTP
Coverage: 17,000+ restaurants · 89 cities · Canada only · UberEats + DoorDash
Why Canuckeats for Agents?
Canadian residents who want AI assistants to handle their food delivery face a gap: UberEats and DoorDash don't offer agent-accessible APIs. Canuckeats bridges that gap. An agent given a task like "order me sushi for dinner" can:
- Find open restaurants near the user's address in their city
- Browse full menus with prices and modifier options
- Get a precise cost breakdown before charging anything
- Place the order and confirm payment in a single tool call
- Track delivery status and report the ETA back to the user
Every order dispatches through real UberEats or DoorDash accounts — the same food, same drivers, same cities.
Get an API Key — Instant Self-Signup
No waiting, no approval queue. Register your agent and start making calls in under 30 seconds:
POST https://mcp.canuckeats.com/register
Content-Type: application/json
{
"agent_name": "MyFoodAgent",
"email": "you@yourproject.com",
"description": "AI assistant that orders lunch for remote teams"
}
Response:
{
"api_key": "your-api-key-shown-once",
"key_id": "uuid",
"scopes": ["catalog:read", "orders:write"],
"medusa_customer_id": "cus_...",
"mcp_endpoint": "https://mcp.canuckeats.com/mcp",
"docs": "https://github.com/canuckeats/MCP",
"message": "Registration successful. Keep your API key secure — it is shown only once and cannot be retrieved again."
}
The key is shown once. Save it immediately. If you lose it, register again with the same email — duplicate email is rejected, so use a unique email per key.
Registration limits: Up to 5 registrations per IP per 24 hours. One key per email address. Disposable email addresses are not accepted.
What Cities Are Covered?
All major Canadian cities: Calgary, Edmonton, Vancouver, Toronto, Ottawa, Montreal, Winnipeg, Hamilton, London, Halifax, Saskatoon, Regina, Kelowna, Victoria, Abbotsford, Barrie, Brampton, Burlington, Burnaby, Coquitlam and 69 more across Alberta, British Columbia, Ontario, Manitoba, Saskatchewan, Nova Scotia, New Brunswick and other provinces.
Use list_cities to get the current full list with restaurant counts.
8 Tools — Full Order Lifecycle
| Tool | Auth | Description |
|---|---|---|
list_cities |
API key | List all 89 cities with restaurant counts |
search_restaurants |
API key | Search by city, cuisine, name, or price range |
get_restaurant |
API key | Full restaurant details |
get_menu |
API key | Complete menu with sections, prices, and modifier options |
validate_order |
API key | Preview exact total — no charge, no order created |
create_order |
API key | Create order + Stripe PaymentIntent (no charge yet) |
confirm_payment |
API key | Charge card and dispatch order to platform |
get_order_status |
API key | Live status, tracking URL, and ETA |
Quickstart: Connect in 60 Seconds
Python (using mcp SDK)
import asyncio, json
from mcp import ClientSession
from mcp.client.streamable_http import streamablehttp_client
async def main():
async with streamablehttp_client(
"https://mcp.canuckeats.com/mcp",
headers={"Authorization": "Bearer YOUR_API_KEY"}
) as (read, write, _):
async with ClientSession(read, write) as session:
await session.initialize()
# Find restaurants
result = await session.call_tool("search_restaurants", {
"city_slug": "vancouver-bc",
"query": "sushi",
"limit": 5
})
restaurants = json.loads(result.content[0].text)
print(f"Found {restaurants['total']} sushi spots in Vancouver")
asyncio.run(main())
TypeScript / Node.js
import { Client } from "@modelcontextprotocol/sdk/client/index.js"
import { StreamableHTTPClientTransport } from "@modelcontextprotocol/sdk/client/streamableHttp.js"
const client = new Client({ name: "my-agent", version: "1.0" }, { capabilities: {} })
await client.connect(new StreamableHTTPClientTransport(
new URL("https://mcp.canuckeats.com/mcp"),
{ requestInit: { headers: { Authorization: "Bearer YOUR_API_KEY" } } }
))
const result = await client.callTool({ name: "list_cities", arguments: {} })
const cities = JSON.parse(result.content[0].text)
console.log(`${cities.length} cities available`)
Claude Desktop Integration
Add to your Claude Desktop MCP config (claude_desktop_config.json):
{
"mcpServers": {
"canuckeats": {
"type": "streamable-http",
"url": "https://mcp.canuckeats.com/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}
Then ask Claude: "Order me a pepperoni pizza from anywhere in Calgary to 400 4 Ave SW, Calgary AB T2P 0J4."
Complete Order Flow
list_cities → get city slug (e.g. "calgary-ab")
↓
search_restaurants → get restaurant id
↓
get_menu → get menu_item_id + modifier options
↓
validate_order → confirm total (no charge)
↓
create_order → get order_id + stripe_payment_intent_id
↓
confirm_payment → charge pm_... and dispatch
↓
get_order_status (poll) → track until delivered
See docs/examples/order-flow.md for a narrated walkthrough with real request/response examples.
Rate Limits
| Endpoint | Limit |
|---|---|
/register |
5 registrations per IP per 24 hours |
| All tools (general) | 60 requests / minute per key |
create_order, confirm_payment |
10 requests / minute per key |
Response headers X-RateLimit-Limit and X-RateLimit-Remaining are included on every MCP response.
Key Facts for Agents
- Canadian addresses only. Province must be a two-letter code:
AB,BC,ON,QC,MB,SK,NS,NB,NL,PE,NT,YT,NU. - Payment method:
confirm_paymentaccepts a Stripepm_*payment method ID. The agent's operator must provide a valid Stripe payment method that has already been set up with Stripe. - Order dispatch is automatic. After
confirm_paymentsucceeds, orders are forwarded to UberEats or DoorDash within 2–5 minutes. Useget_order_statusto track. - Delivery platform is chosen automatically based on which platform has that restaurant in its catalog.
source_idon menu items is used internally — you do not need to reference it directly. Just usemenu_item_id(the integeridfield) when callingcreate_order.- Modifiers are optional. If a menu item has required modifiers (e.g. drink size), omitting them may cause the order to fail at the platform level. Check
modifier_groups[].min_selectionsto know which are required. - Scheduled delivery is supported via the
scheduled_forISO 8601 field oncreate_order. Use ASAP delivery by omitting this field.
Supported Delivery Platforms
| Platform | Coverage |
|---|---|
| UberEats | All 89 cities |
| DoorDash | All 89 cities |
Orders are placed on whichever platform has the requested restaurant. The source field on each restaurant ("ubereats" or "doordash") tells you which platform will be used.
Canadian Cities — Full List
Alberta · British Columbia · Manitoba · New Brunswick · Newfoundland · Nova Scotia · Ontario · Prince Edward Island · Quebec · Saskatchewan
Use list_cities for the live list. Notable cities include:
Alberta: Calgary, Edmonton, Red Deer, Lethbridge, Airdrie, Medicine Hat, Grande Prairie, Fort McMurray
BC: Vancouver, Surrey, Burnaby, Richmond, Kelowna, Abbotsford, Victoria, Coquitlam, Langley
Ontario: Toronto, Ottawa, Hamilton, London, Brampton, Mississauga, Barrie, Windsor, Kitchener, Ajax, Aurora
Quebec: Montreal (selected areas)
Prairie & Atlantic: Saskatoon, Regina, Winnipeg, Halifax, Moncton, Fredericton
API Reference
- Full tool schemas with TypeScript types
- Quick start guide with code samples
- End-to-end order walkthrough
- Machine-readable tools manifest
Support
- Issues / questions: Open a GitHub issue in this repo
- API key lost: Re-register with a different email, or email hello@canuckeats.com
- Order issues: Email hello@canuckeats.com with your
order_id - Website: canuckeats.com
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.