Skip to main content
Glama
market-screening

market-screening

by 3vgen

market-screening

MCP server that helps an LLM answer "where should I buy my grocery list?" across Russian retail chains. You give the assistant a shopping list; it searches the products in the stores you shop at, matches comparable items (by price per kg / l), and computes the cheapest plan: everything in one store, or split across 2–3 stores.

Status: early prototype.

Chain

Status

Source

Notes

Magnit

✅

website JSON API

per-store prices, stock, promos, card-only badges

VkusVill

✅

server-rendered search page

default-region (Moscow) prices, no per-store stock

Dixy

❌

—

listing endpoint behind WAF (403)

Pyaterochka, Perekrestok, Lenta

❌

—

JS anti-bot challenge

How it works

Tool

What it does

list_stores

Stores from your config.toml

search_products(query, stores?, limit?)

Products with price, promo/old price, price per kg/l, stock, card-only badges

optimize_basket(basket, max_stores, extra_store_cost)

Exact totals: best single store vs best split, saving

The LLM does the fuzzy part (which "молоко" in each store is comparable); the server does the arithmetic, so totals are not hallucinated.

Data comes from the same public JSON endpoints the chains' websites use, without login. Results are cached for 15 minutes to keep the load on the sites minimal.

Related MCP server: PriceAtlas MCP Server

Setup

python3 -m venv .venv
.venv/bin/pip install -e ".[dev]"
cp config.example.toml config.toml   # put your store codes there

Finding your Magnit store code: open magnit.ru, pick your store, then DevTools → Network → goods/search → Payload → storeCode.

Claude Code

claude mcp add market-screening -e MARKET_SCREENING_CONFIG=$PWD/config.toml -- $PWD/.venv/bin/market-screening

Claude Desktop / other MCP clients

{
  "mcpServers": {
    "market-screening": {
      "command": "/path/to/market-screening/.venv/bin/market-screening",
      "env": { "MARKET_SCREENING_CONFIG": "/path/to/market-screening/config.toml" }
    }
  }
}

Network note

Chains often block foreign IPs. If you use a VPN, route the chains' domains directly (split tunneling), e.g. for v2rayA RoutingA — place these above any geosite: rules:

domain(domain:magnit.ru, domain:dixy.ru, domain:vkusvill.ru)->direct

Adding a chain

Create src/market_screening/adapters/<chain>.py with an Adapter subclass that implements _search(store, query, limit) -> list[Product], register it in adapters/__init__.py, and add a parsing test with a small fixture in tests/.

Disclaimer

Unofficial project, not affiliated with any retailer. Endpoints are undocumented and may change. Intended for personal use; respect the sites' terms and keep request volume low.

License

MIT

Available Tools

3 tools
list_storesC

Stores configured by the user (chain, store code, name). Use the key field elsewhere.

Inputs: no named parameters. Returns: result

optimize_basketA

Compute the cheapest way to buy the basket.

basket: [{"item": "молоко", "quantity": 2, "offers": {"magnit:123456": 89.99, "dixy:123": 92.5}}]

  • offers: price in rubles per one unit of quantity, for the product you matched in each store; omit a store if it has no suitable product. max_stores: how many different stores the user is willing to visit. extra_store_cost: rubles the user is willing to pay to avoid each extra store visit.

Inputs: basket (required), max_stores, extra_store_cost.

search_productsA

Search a product in the configured stores.

query: what to look for, in Russian, e.g. "молоко 2.5%", "куриное филе". stores: optional list of store keys from list_stores (default: all). Returns per store a list of products: price in rubles, old_price if on promo, unit_price per kg / l / pcs (compare by this), stock, badges, url. Search is fuzzy: results can include unrelated items (e.g. oat drink for "молоко").

Inputs: query (required), limit, stores.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 3 tools

    Disambiguation5/5

    Each tool occupies a distinct stage of a clear pipeline: list_stores enumerates configured stores, search_products finds items with prices, optimize_basket computes the cheapest multi-store purchase. There is no overlap in purpose or output, so an agent cannot plausibly misselect.

    Naming Consistency5/5

    All three names follow the same snake_case verb_noun convention (list_stores, search_products, optimize_basket), with the verb accurately signaling the operation (list vs. search vs. optimize).

    Tool Count4/5

    Three tools map cleanly onto the domain's three steps (enumerate stores, search products, optimize basket), so no tool is filler. It sits at the lean end of appropriate: there is no direct product-lookup or price-refresh tool, which would round the set out.

    Completeness4/5

    The end-to-end flow (stores -> products -> cheapest basket) is covered and the schemas are explicit about how data passes between tools. The gap is that prices/offers must be hand-carried from search_products into optimize_basket, and there is no key-based product or price lookup to refresh an item without a new fuzzy search.

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      Not graded
      maintenance
      Enables AI assistants to track global food prices, search products by barcode or name, and compare costs across 27 countries. It provides tools for real-time price scraping and data aggregation from major international supermarket chains.
      8
      4
      MIT
    • A
      license
      C
      quality
      D
      maintenance
      Enables AI assistants and MCP clients to search Pyaterochka stores, browse categories, find products with filters for price, brand, and promotions, and retrieve product details, store hours, and promotion conditions through the exposed MCP tools.
      12
      10
      MIT