Grubhub

io.github.markswendsen-codePublished by GitHub account markswendsen-codev0.1.3Updated Oct 28, 2026

MCP server for GrubHub - let AI agents order food and manage restaurant delivery autonomously

VerifiedSTDIODesktop onlyLocation & LifestyleBrowser Automation

Overview

AI-generated overview

Grubhub connector that lets an AI assistant log in, search restaurants, build a cart, place food delivery orders and track them.

What you can do
  • Log in, check login status and clear the stored session
  • Set a delivery address and search restaurants by name, cuisine or food type
  • Browse restaurant details and full menus with item prices
  • Add items to the cart, view it and clear it
  • Preview or place an order at checkout
  • Track an active delivery and browse past order history
When to use it
Useful when you want an assistant to handle Grubhub food delivery end to end: finding restaurants near an address, assembling an order from a menu, placing it and following the delivery status, without switching to the app yourself.
Requirements
Runs locally as an npm package (npx @striderlabs/mcp-grubhub) over stdio, desktop only. Needs a Grubhub account with GRUBHUB_EMAIL and GRUBHUB_PASSWORD set, plus network access; login uses Playwright browser automation and saves session cookies to a local config file.
Before you install
The server stores your Grubhub account password in GRUBHUB_PASSWORD and keeps session cookies on disk, so anyone with access to that machine or config file can act as you. Checkout can place real orders that spend money, so review the preview step before confirming. Cart and order actions change your account state.

Installation

In SourceWeft

  1. Open Grubhub in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

@striderlabs/mcp-grubhub

Order food delivery via Grubhub using AI agents

[npm] [MCP Registry] [License: MIT]

Part of Strider Labs — action execution for personal AI agents.

For Agents

This connector is designed for personal AI agents to autonomously order food on behalf of their human. Agents can:

  • Understand context: "My user usually orders from Thai places in the morning and Italian restaurants in the evening"
  • Optimize for preferences: Search restaurants that match dietary restrictions, price range, or past favorites
  • Execute autonomously: Place complete orders (search → menu → cart → checkout) in a single agent turn
  • Track delivery: Monitor orders and notify the user when food is arriving

Example agent workflow:

Agent: "The user said they're hungry. Search for Thai restaurants near 123 Main St, find Pad Thai, add a large order to the cart, and place it with default payment method."

The agent calls:

  1. grubhub_set_address → "123 Main St"
  2. grubhub_search_restaurants → { query: "Thai", cuisine: "Thai" }
  3. grubhub_get_restaurant → Browse menus
  4. grubhub_add_to_cart → Add Pad Thai
  5. grubhub_checkout → { confirm: true } to place order
  6. grubhub_track_order → Monitor delivery status

All in seconds, without human intervention.

Features

  • 🔐 Login with email/password via browser automation
  • 🔍 Search restaurants by name, cuisine, or food type
  • 📜 Browse menus with full item details and prices
  • 🛒 Manage cart — add items, view, and clear
  • 💳 Place orders with a confirmation step
  • 📍 Track orders with real-time status
  • 📋 Order history — browse past orders
  • 💾 Persistent sessions — stay logged in across restarts

Quick Start

Install

bash
npm install @striderlabs/mcp-grubhub

Or with npx:

bash
npx @striderlabs/mcp-grubhub

Claude Desktop Configuration

Add to your ~/Library/Application Support/Claude/claude_desktop_config.json:

json
{  "mcpServers": {    "grubhub": {      "command": "npx",      "args": ["-y", "@striderlabs/mcp-grubhub"]    }  }}

Your Agent Can Now

"Order Thai food from nearby restaurants with delivery in under 30 minutes"→ Agent searches → Browses menus → Places order → Confirms delivery

Metrics

  • Weekly downloads: 123 (Apr 10-17, 2026) — Growing Strider Labs connector
  • Status: ✅ Live in production
  • Reliability: 85%+ task completion rate
  • Discovery: npm, mcpservers.org, ClawHub, PulseMCP, Glama

Authentication

On first use, call grubhub_login with your email and password. The server uses Playwright browser automation to log in and saves session cookies to ~/.config/striderlabs-mcp-grubhub/cookies.json for future use.

To log out:

grubhub_logout

Available Tools

Authentication

ToolDescription
grubhub_statusCheck login status
grubhub_loginLog in with email and password
grubhub_logoutClear stored session

Ordering

ToolDescription
grubhub_set_addressSet delivery address
grubhub_search_restaurantsSearch by query or cuisine
grubhub_get_restaurantGet restaurant details and full menu
grubhub_add_to_cartAdd item to cart
grubhub_view_cartView current cart
grubhub_clear_cartClear all cart items
grubhub_checkoutPreview or place order

Tracking

ToolDescription
grubhub_track_orderTrack active order status
grubhub_get_ordersGet order history

Example Usage

1. grubhub_login — log in with your credentials2. grubhub_set_address — set your delivery address3. grubhub_search_restaurants — find a restaurant4. grubhub_get_restaurant — browse the menu5. grubhub_add_to_cart — add items6. grubhub_checkout (confirm=false) — review order7. grubhub_checkout (confirm=true) — place order8. grubhub_track_order — track delivery

Available Elsewhere

License

MIT — Strider Labs

Source: README.md at commit c014fb1

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.1.3LatestSep 16, 2026

Score

40% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

    Declare a license in the repository or the registry entry.