How to Connect MPSTATS to Cursor via MCP

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → MPSTATS, paste your credentials, mint an MCP API key and you're done. No Docker, no git clone, no local server.

MPSTATS + Cursor

MPSTATS marketplace analytics for Wildberries and Ozon: estimated sales and revenue, prices and stock per warehouse and size for any item, search queries and positions, competitors of an item, and figures and trends of niches, categories, brands and sellers. Read-only.

Prerequisites

See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:

MPSTATS_API_TOKEN

Step 1 — Get credentials

Setting up takes about 2 minutes. You need an MPSTATS account whose plan includes API requests.

1. Sign in at https://mpstats.io, open the account settings and copy the token from the API token block into MPSTATS_API_TOKEN.
2. Install the connector and run mpstats_wb_get_keyword_frequency with a query such as джинсы: it returns a weekly series and confirms the token.

The token is sent in the X-Mpstats-TOKEN header. It changes when the account password is changed or Change API token is pressed; paste the new one then.

Market data, not your seller cabinet: MPSTATS collects public data from Wildberries and Ozon, so the connector reads any item, niche, category, brand or seller, yours or a competitor's. It does not reach your own seller cabinet: for your orders, finances and stock use the Wildberries or Ozon connector.

Identifiers: items are marketplace SKUs, the Wildberries article number (nmId) or the Ozon SKU. When you only have a name, the AI resolves it first with mpstats_wb_find_items or mpstats_oz_find_items (SKUs or a keyword), mpstats_wb_search_subjects, mpstats_oz_search_niches, mpstats_wb_search_sellers, mpstats_oz_search_sellers or mpstats_wb_search_brands. Wildberries subjects and Ozon niches and sellers use numeric ids; categories are full paths with /, exactly as MPSTATS shows them (for example Женщинам/Джинсы); brands are names.

What it reads on Wildberries:

• One item: the full card (brand, seller, rating, FBO and FBS stock, orders, price and discount, sizes, photos, first seen date); one row per day with sales, estimated revenue, stock, price and discounts, rating, comments, average category position and visibility, optionally with search positions and traffic; a search summary compared with the previous period (number of queries, average organic and ad position, visibility, estimated traffic); the search queries it ranks for, with frequency and its organic and ad position per day; stock and sales by warehouse and by size, to spot sold-out sizes.
• Competitors of an item: what Wildberries shows as similar on the item page, items whose similar block shows this SKU, items Wildberries marks as identical, and similar items matched by MPSTATS, each with sales, revenue, price and stock.
• Niches (subjects) and categories: their items with sales, revenue, price, stock, rating and position; a split by subcategories, subjects, brands or sellers; and market figures per day, week, month or quarter (sales, revenue, average price, stock, items, brands and sellers with sales) for trend and seasonality.
• Sellers and brands: their items with sales, revenue, price and stock. Seller search also returns the INN.
• Search demand: how often a query is searched on Wildberries, as a weekly series.

What it reads on Ozon: item lookup and the full card; daily sales, revenue, stock, price, rating and comments; search queries with frequency and positions; positions in each category per day; stock per warehouse on one day (FBO, optionally FBS); sales by warehouse; similar items matched by MPSTATS. Niches and their items; categories with their items, a split by subcategories, niches, brands or sellers, and the trend per day, week, month or quarter; sellers (with OGRN) and brands with their items.

Periods and figures: d1 and d2 are YYYY-MM-DD, and the AI always passes them because the default period differs from method to method. Data runs to yesterday. Sales and revenue are MPSTATS estimates from public marketplace data, in roubles; treat them as estimates, not as the seller's accounting. fbs=1 adds sales shipped from the seller's own warehouse (FBS), 0 counts marketplace warehouses (FBO) only.

Paging, sorting and filters: list tools return startRow, endRow, total and data, at most 500 rows per call; the AI pages by moving startRow and endRow. sortModel sorts, for example [{"colId": "revenue", "sort": "desc"}], and filterModel filters columns, for example revenue greater than 100,000 (text filters: equals, contains, startsWith; number and date filters: lessThan, greaterThan, inRange).

Read-only: every tool only reads, so there is nothing to confirm and no tool is switched off at install.

Cost and limits: every call counts against the request quota of your MPSTATS plan, the test call included. The AI keeps periods short and pages small, and reuses what it already fetched instead of calling again.

Common errors: 401 means a wrong or changed token. 429 means the plan's request quota is used up; the message says why and Retry-After says when to retry. 202 means MPSTATS is still preparing the report: call again a little later. 500 carries the reason in message.

The connector changes nothing on Wildberries or Ozon: it does not edit cards, prices or ads, and it does not read your orders, payouts or seller reports.

Step 2 — Install the adapter

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Step 3 — Add the connector in Cursor

Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide shows YOUR_SERVER_ID.

Cursor reads MCP servers from ~/.cursor/mcp.json. Add this entry:

{
  "mcpServers": {
    "anythingmcp": {
      "url": "https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID",
      "headers": { "X-API-Key": "YOUR_MCP_API_KEY" }
    }
  }
}

1. Create an MCP API key in AnythingMCP under MCP Servers → your server → API keys, and paste it in place of YOUR_MCP_API_KEY.
2. Save the file and restart Cursor.
3. Open Cursor → Settings → MCP and check that anythingmcp is listed and connected.
4. Start chatting: the agent can now call the MPSTATS tools.

Available tools

Example prompts

• "How many units did Wildberries item 123456789 sell per day over the last 30 days, and when did it run out of stock?"
• "Which sizes of this Wildberries item are sold out, and in which warehouses is stock left?"
• "Show the top 20 items of the Wildberries niche "Джинсы" by revenue last month, with price and seller."
• "Which search queries does Ozon SKU 987654321 rank for, and at which positions?"
• "Show the monthly revenue of the Wildberries category Женщинам/Платья over the last year: when is the season?"

FAQ

Does Cursor support custom MCP servers on the free tier? Yes, MCP is available on Hobby, Pro and Business plans.

Next steps

• Connector landing page
• Connect to Claude
• Connect to ChatGPT
• Connect to Gemini
• Connect to Meta Muse
• Connect to Microsoft Copilot