We Built an MCP Server for Real Discounts and Promo Codes — Now Listed in the Official Registry

Description and use

A public MCP server exposes real discount data to AI assistants, registered in the official MCP Registry.

Problem

AI assistants need to search real shopping offers without scraping websites or relying on invented discounts.

Outcome

The MCP server allows AI assistants to retrieve real discount offers, improving shopping discovery and enabling developers to build commerce agents without scraping.

Workflow

• 1. Implement MCP server exposing 8 read‑only tools (search_deals, find_promo_codes, get_store_offers, get_best_deals, list_stores, list_categories, get_offer_details, compare_deals).
• 2. Host server at https://mcp.discounthub.uz/mcp with streamable HTTP transport.
• 3. Register server in MCP Registry under uz.discounthub/deals-and-coupons (v1.0.0).
• 4. Test integration by connecting to GitHub Copilot and issuing natural‑language queries (e.g., “Find a laptop with at least 30% off”).
• 5. Validate returned offers contain real product data, price, discount, and shopping link.
• 6. Perform regression tests and improve store‑name resolution to handle regional merchant names.

Evidence

We've launched a public Model Context Protocol (MCP) server that connects compatible AI agents to our existing discounts and promotions catalogue.

Source

Our server exposes 8 read-only tools: search_deals, find_promo_codes, get_store_offers, get_best_deals, list_stores, list_categories, get_offer_details, compare_deals.

Source

We connected DiscountHub MCP to GitHub Copilot and ran several natural-language tests.

Source

During testing, we found that regional merchant names could cause incomplete search results. We improved store-name resolution and added regression tests.

Source

DiscountHub is now registered as: uz.discounthub/deals-and-coupons Version: 1.0.0

Source