- Home
- Skills
- E-commerce
- Wildberries Competitor & Seller Analyst
Wildberries Competitor & Seller Analyst
Turn your AI agent into an analyst for **Wildberries**, Russia's largest marketplace.
Free
Works with the AI Agents you already use
Wildberries Competitor & Seller Analyst
Example session with this skill installed
Here are three wireless headphones on Wildberries: 1470151551 (mine), 1453309500 and 1337936355. Compare them, tell me who changed prices recently, and what buyers complain about in the competitors' reviews.
- Read your context and instructions
- Compiled the wildberries competitor
Real output: rf-marketplaces-mcp v1.1.0 tools wb_compare, wb_price_watch and wb_review_insights called on 23 Oct 2026 for these articles, written up the way the agent answers. Your numbers will differ when you run it.
3 products compared — data as of 23 Oct 2026
| Product (article) | Seller | Last known price | Rating (reviews) | Stock / delivery |
|---|---|---|---|---|
| Lirano Pro 2 earbuds (1470151551) — yours | Lirano official store | 623 ₽ (13 Sep) | 4.9 (7 611) | n/a |
| Yandex Drops TWS with Alice (1453309500) | Modern Device | 6 412 ₽ (20 Sep) | 4.9 (183) | n/a |
| Ural MOLOT H130 over-ear (1337936355) | ASTRONIUM official store | 4 936 ₽ (30 Aug) | 5.0 (27) | n/a |
Wildberries currently doesn't publish current prices, discounts, stock or delivery times; prices shown are the last known ones from WB's price history, with their dates.
Leaders: lowest last known price — yours; best rating — MOLOT H130 (5.0, but only 27 reviews); most reviews — yours. No stock or delivery leader can be named.
Price changes (4 weeks up to each product's last known price) — WB has published no price points after 20 Sep for these products, so I can't tell who changed prices in October.
- Yours: −22.2% (800 → 623 ₽ between 30 Aug and 13 Sep), at its historical minimum.
- Yandex Drops: +15.8% in the 4 weeks to 20 Sep (history range 5 540–6 426 ₽).
- MOLOT H130: −1.3%, at its historical minimum (4 936 ₽ on 30 Aug).
For your own current price, connect your seller token and I'll check wb_seller_prices.
What buyers complain about in the competitors' reviews (1–3★)
- Yandex Drops: 3.5% negative (6 of 183 by WB's star distribution); the seller answered 100% of the reviews WB returns. The complaints are about sound for the price — "too much treble, zero bass" («Басов - 0!», 3★, 14 Oct), "sound quality unexpectedly poor" (2★, 9 Oct) — plus the Alice assistant behaving unpredictably (3★, 7 Oct). Recurring words: «алиса» (Alice, 3 reviews), «плохо» (bad), «брак» (defect), «поддержка» (support), «слышит» (hears) — 2 each. WB's own tag "Quality": 41 positive, 2 negative.
- MOLOT H130: no 1–3★ reviews at all (rating 5.0 over 27 reviews); the seller answered 100% of reviews.
Opportunities: both competitors answer every review — match that on your card. The Yandex Drops cost ten times more than your last known price and still draw "not worth the money" sound complaints, so if your sound holds up, make sound quality for the price the centre of your card.
Connects securely to your tools. The creator never sees your data.
About this skill
Turn your AI agent into an analyst for Wildberries, Russia's largest marketplace. Give it article numbers or wildberries.ru links and ask in plain words:
- "Compare my product with these 5 competitors" — seller, rating, number of reviews and the last known price with its date, plus the leader in each available metric.
- "Who changed prices?" — weekly price history for up to 20 products with dated signals: historical minimum or maximum, a 10%+ drop or rise, a likely sale.
- "What do buyers complain about?" — negative reviews (1–3★) of up to 10 products mined for recurring complaints, plus the share of negative reviews, how often sellers answer, WB's own review tags and fresh examples.
- "Audit my product card" — photos, video, rich content, description length, filled characteristics, rating, reviews, negative share, seller answers and last known price against the competitors' median, with a prioritised fix list.
No account or token is needed for any of that. The server reads the public product files, price history and reviews the Wildberries website itself loads.
What Wildberries no longer shows (October 2026). WB closed its product-card API to external requests, so competitors' current price, discount, stock, sizes and delivery time are not available. The skill teaches the agent to say so plainly, mark those values "n/a", show the last known price from WB's price history with its date instead, and never guess a missing number. Your own current prices are still available through your seller token.
Optional seller mode. Add your official Wildberries seller API token (WB_API_TOKEN) and the agent can also report sales, returns and payout, orders and cancellations, warehouse stock (out of stock / running low), your current prices and discounts, and list unanswered reviews and questions. It can draft answers in Russian and publish them — only after you approve the exact text.
Language and market. Wildberries is a Russian marketplace: prices are in rubles, product data and reviews are in Russian. The agent answers in your language and translates review quotes and complaint themes for you.
Honest limits. Products are looked up by article number only (WB closed keyword search and the catalog to external requests). Price history is weekly, not published for every product, and has updated irregularly since September 2026, so the last known price can be weeks old. The public data is not an official API and WB can change it; the server does not bypass WB's anti-bot protection. WB returns only the latest ~1000 reviews per product. Seller statistics are limited by WB to 1 request per minute.
Works with Claude Code, Claude Desktop, Cursor, VS Code Copilot, Windsurf, Codex and any agent that supports local MCP servers. Built on the open-source rf-marketplaces-mcp server (MIT), which runs on your machine. Setup takes about 3 minutes; ready-made configs and step-by-step instructions are included.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for E-commerce.
- Lookup by article number only; no keyword or category search.
- Competitors' current price, discount, stock, sizes, delivery time and seller rating are unavailable while Wildberries keeps its card API closed (since October 2026); they are reported as "n/a", with the last known price and its date instead.
- Price history is weekly, not published for every product, and has updated irregularly since September 2026 — the last known price can be weeks old or missing. Continuous monitoring means re-running the tool (the server stores nothing between runs).
- Public product data comes from unofficial Wildberries endpoints, not a documented API: they can change without notice. The server does not bypass WB's anti-bot protection; networks that cannot reach the WB CDN (some VPNs and data-centre IPs) get "not found" or errors.
- Only the latest ~1000 reviews per product are available; star distribution and negative share cover all reviews, complaint texts only the loaded ones. Complaint extraction is word/phrase frequency, interpreted by the agent.
- Seller statistics are rate-limited by WB to 1 request per minute per method; stock data refreshes about every 30 minutes; seller prices exclude WB's SPP and promotions.
- Wildberries only; Ozon and Yandex Market are not supported yet.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 25 days ago
- Free with an Agensi account
- Accounts
GitHub
- Software
- rf-marketplaces MCP
Node.js 18+
- Optional
- Wildberries seller API token
Run the MCP over stdio with npx github:penmadebykisss/rf-marketplaces-mcp#v1.1.0. The seller token is only for wb_seller_* tools.
Permissions
Network access: the locally running MCP server makes HTTPS requests to public Wildberries hosts (card.wb.ru — tried first, currently closed by WB; basket-NN.wbbasket.ru; feedbacks1/2.wb.ru; feedback-bt.wildberries.ru; feedback-view-*.wb.ru) to read product, price-history and review data, and — only if `WB_API_TOKEN` is set — to the official Wildberries seller API (common-api, statistics-api, feedbacks-api, discounts-prices-api on wildberries.ru). It does not bypass WB's anti-bot protection. The skill contains no code and no secrets; the token lives only in the user's MCP config or environment. Two tools (answering reviews and questions) publish text on Wildberries; the skill requires explicit user approval of the exact text before each one. Everything else is read-only.