Try BuyWhere MCP Shopping Agent — Interactive Claude Artifact + ChatGPT GPT Template

#ai #mcp #shopping #agents

Want to see how an AI shopping agent actually works — without writing any code?

We built two things:

A Claude Artifact — an interactive HTML/JS shopping comparison agent powered by BuyWhere MCP
A ChatGPT GPT template — ready-to-use config for a custom GPT that compares prices across Singapore retailers

Both work out of the box. Fork the artifact, tweak the GPT config, or just play with the demo to see what MCP-powered commerce looks like.

What You Get

Claude Artifact

Open the artifact in Claude and you'll see a full shopping agent UI:

Search 3M+ products across Shopee, Lazada, Amazon SG, Cold Storage, Guardian, Giant, and 7-Eleven
Interactive price comparison charts
Side-by-side retailer pricing with savings callouts
The actual MCP config shown in the sidebar so devs can copy-paste

Fork it here: BuyWhere MCP Shopping Agent Artifact

ChatGPT GPT Template

A pre-built GPT configuration that knows how to use BuyWhere's MCP tools:

Pre-configured with all 5 MCP tools
Built-in conversation starters
Retailer and regional coverage baked into the knowledge base
Drop-in JSON config for your own GPT

Download the config: buywhere-gpt-config.json

How the Artifact Works

The HTML artifact simulates what happens when Claude calls BuyWhere MCP tools:

search_products — finds matching products across the catalog
compare_products — compares prices across retailers
Results normalized from a unified product layer (just like the real BuyWhere MCP)

The sidebar shows the actual claude_desktop_config.json snippet so developers can go from "try it" to "run it" in 30 seconds.

Get Started for Real

To use BuyWhere MCP in your own agents:

npm install @buywhere/mcp-server

Add to Claude Desktop:

{
  "mcpServers": {
    "buywhere": {
      "command": "npx",
      "args": ["-y", "@buywhere/mcp-server"],
      "env": {
        "BUYWHERE_API_KEY": "bw_free_your_key"
      }
    }
  }
}