Wish Scraper MCP Server
Pricing
Pay per event + usage
Wish Scraper MCP Server
MCP server + scraper for Wish.com. Let Claude, ChatGPT & Cursor search Wish products, get detail, reviews, categories & autocomplete live via Model Context Protocol — or run it as a classic scraper. Fast direct API. Pay per tool call.
Pricing
Pay per event + usage
Rating
0.0
(0)
Developer
axly
Maintained by CommunityActor stats
0
Bookmarked
1
Total users
0
Monthly active users
a day ago
Last modified
Categories
Share
Give your AI assistant live access to Wish.com. This actor runs as a Model Context Protocol (MCP) server so Claude, ChatGPT, Cursor, n8n, and other agents can search Wish products, pull full detail and reviews, browse categories, and get autocomplete suggestions — in conversation, no code. It also runs as a classic scraper for scheduled dataset jobs.
Who it's for
- AI-assistant users — ask Claude/ChatGPT/Cursor for Wish products, prices, and reviews and get structured answers instantly.
- Agent & automation builders — a hosted MCP tool for Wish in LangChain, n8n, or custom agents.
- Developers — one endpoint for Wish product search + detail + reviews.
MCP tools
| Tool | What it does |
|---|---|
search_wish_products(query, limit, fetch_detail) | Search Wish by keyword; optional detail enrichment |
get_wish_product(product_id_or_url, with_reviews) | Full product detail: price, variations, images, sold-out |
get_wish_reviews(product_id_or_url, limit) | Customer reviews for a product |
wish_autocomplete(query) | Search-box suggestions for a partial query |
get_wish_categories() | Category tree (names + handles) |
browse_wish_category(category_handle, limit) | Products within a category |
Connect it
Run the actor in Standby mode and point your MCP client at the server URL:
Example Claude Desktop / Cursor config (Streamable HTTP):
Then ask: "Search Wish for wireless earbuds under $10 and show ratings" or "Get the detail and reviews for this Wish product URL."
Output fields
Product tools return normalized rows: product_id, name, url, image_url,
price_usd, price_local, currency, crossed_price_usd, rating,
review_count, num_bought, merchant_id, merchant_name, is_sold_out,
variation_count, images, and (on request) reviews.
Classic run mode
Prefer a one-shot dataset job? Provide searchQueries / productIds in the input
and run it like any scraper — rows land in the dataset. See the companion
Wish Product Scraper actor for the full multi-mode dataset scraper (category,
collection, bulk IDs, reviews).
FAQ
What is MCP? The Model Context Protocol — an open standard that lets AI assistants call external tools. This actor is an MCP server for Wish.com.
Do I need an account or proxies? No. It calls Wish's public mobile API directly and works out of the box.
How fresh is the data? Live — every tool call fetches current data.
Which assistants work? Any MCP-capable client: Claude Desktop, Cursor, ChatGPT (via connectors), n8n, LangChain, and custom agents.
Is it reliable? It calls the API directly (no fragile browser automation) and serializes calls safely; per-item failures are skipped and logged.
