Skip to main content
Glama
gobuy-ai

GoBuy Product Trust MCP Server

Official
by gobuy-ai

GoBuy Product Trust MCP Server

A public, read-only MCP server that gives AI agents GoBuy marketplace-evidence scores. One call returns a 0–100 evidence index, observed and missing signals, freshness, and product context.

Evidence Scores describe available marketplace records, not product quality, safety, value, suitability, seller legitimacy, or a purchase recommendation. Retailers do not endorse GoBuy.

Public endpoint: https://gobuy.ai/mcp

Service statistics are available at https://gobuy.ai/stats.

Tools

  • check_product_trust: inspect one Amazon, Walmart, Target, or Best Buy listing.

  • compare_products: inspect and rank 2–20 listings by Evidence Score.

Both tools accept native retailer IDs. Example:

{
  "retailer": "amazon",
  "product_id": "B0BXYCS74H"
}

Comparison input:

{
  "products": [
    { "retailer": "amazon", "product_id": "B0BXYCS74H" },
    { "retailer": "walmart", "product_id": "123456789" }
  ]
}

Run locally

Requirements: Node.js 20+ and pnpm.

cp .env.example .env
# Set INTERNAL_API_URL to the GoBuy API origin.
pnpm install
pnpm build
PORT=3000 INTERNAL_API_URL=https://your-gobuy-api.example pnpm start

Streamable HTTP is available at http://localhost:3000/mcp. GET /mcp without an SSE Accept header is a health check. Cumulative in-process tool counts are at GET /stats.

No client authentication is required in v1. Requests are limited per IP to 60/minute and 1,000/day. Limit responses use HTTP 429 and Retry-After.

Client configuration

Claude Desktop

Claude Desktop commonly uses a local bridge for remote Streamable HTTP servers:

{
  "mcpServers": {
    "gobuy-product-trust": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://gobuy.ai/mcp"]
    }
  }
}

Cursor

{
  "mcpServers": {
    "gobuy-product-trust": {
      "url": "https://gobuy.ai/mcp"
    }
  }
}

Generic Streamable HTTP client

Point the client at https://gobuy.ai/mcp; no authorization header is needed.

Inspector

Run the official MCP Inspector:

npx @modelcontextprotocol/inspector

Choose Streamable HTTP, enter https://gobuy.ai/mcp, connect, list tools, then call check_product_trust or compare_products.

Docker

INTERNAL_API_URL=https://your-gobuy-api.example docker compose up --build

The multi-stage image runs as the unprivileged node user and includes a /mcp healthcheck.

Operations

Each tool call emits one JSON log line with timestamp, tool, latency_ms, and status. /stats counters reset when the process restarts.

License

MIT

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/gobuy-ai/gobuy-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server