Doriandarko

Doriandarko / webmcp-starter

Public

WebMCP starter demo — a DoorDash-style food delivery app with 9 AI agent tools (imperative + declarative)

195
16
100% credibility
Found Feb 17, 2026 at 170 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
HTML
AI Summary

A single-file web demo of a food delivery app that AI assistants can use to search restaurants, build carts, apply discounts, and check out.

How It Works

1
🔍 Discover the Food Demo

You hear about a fun demo where AI helps order food like on a delivery app, just by chatting.

2
🛠️ Prepare Your Browser

Get a special test version of your browser, turn on a new feature, and add a helpful add-on to connect AI to websites.

3
🍜 Open the Food App

Find the app file on your computer and open it in the test browser to see a yummy restaurant menu.

4
💬 Chat with the AI Helper

Click the add-on, type what you're craving like Italian pizza and dessert, and watch the AI get to work.

5
🛒 Build Your Order

The AI searches eateries, picks your favorites, adds them to the basket, applies a welcome discount, and tallies it up.

6
📱 Finish Checkout

Give your address and a tip, and the AI fills out the payment form to complete the order smoothly.

✅ Order Placed!

Your pretend meal is ready to 'deliver'—you've experienced AI making online shopping feel effortless and magical.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 170 to 195 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is webmcp-starter?

WebMCP-starter is a single-file HTML demo of a DoorDash-style food delivery app that exposes 9 WebMCP tools for AI agents—8 imperative ones for searching restaurants, browsing menus, managing carts, applying promo codes, and tracking orders, plus 1 declarative checkout form. It solves the chicken-and-egg problem of testing browser-native AI agent tools by giving you a ready-to-run app where agents like Gemini can handle a full ordering flow via natural language prompts. Open it in Chrome Canary, and you're simulating real delivery app interactions in minutes.

Why is it gaining traction?

Its 159 stars come from devs excited about WebMCP, the proposed standard turning browsers into agent-friendly environments with structured tools—no custom APIs needed. What hooks users is the complete end-to-end demo: crave Italian, add pizza and tiramisu, promo, checkout— all via agent prompts, blending imperative actions with declarative HTML forms. It's a low-friction way to grok agent apps versus verbose SDKs or mock servers.

Who should use this?

Frontend devs prototyping AI agent tools for e-commerce or delivery apps. AI engineers experimenting with browser agents on real flows like cart management and order status. Teams scouting WebMCP for declarative UIs in agent-driven web experiences.

Verdict

Worth forking as a WebMCP starter demo if AI agents intrigue you—solid docs and test prompts make it approachable, despite 1.0% credibility score, 159 stars, and early-stage flags in Chrome Canary. It's a raw proof-of-concept; polish it for anything beyond demos.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.