ecommerce-scraper
Skill0 starsChanged 41 days agoMITView on GitHub
Scrape product pages and search marketplace listings through ToolRouter's MCP server - product title, price, variants, images and specs from any store URL (Shopify detected automatically), plus listings from eBay, Facebook Marketplace, Vinted, Gumtree, Craigslist, Mercado Libre, OLX, Rakuten, Carousell, Trade Me and Google Shopping. Use for competitor pricing, product research, resale price checks and listing monitoring.
In this repository7 other files it gives its agents.
- ad-library-scraperSkill
- bot-store-marketplaceSkill
- magicscreenshotsSkill
- sentrydockSkill
- social-media-scraperSkill
- toolrouterSkill
- tradehandSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 46
Call these through use_tool with tool, skill and input.
- Very specific: line 52
Sourcing: search by model number across regions, then open the cheapest listings with listing_details.
- Very specific: line 53
Monitoring: run watchlist_snapshot on a schedule and diff against the last snapshot the user saved.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
E-commerce Scraper
Turn store pages and marketplace searches into structured product data from one MCP connection. No scraper to host and no marketplace developer accounts.
Installation
Install with pnpm dlx skills add Humanleap/agent-skills --skill ecommerce-scraper. Connect the client's remote MCP server to https://api.toolrouter.com/mcp (Claude: Customize → Connectors → Add custom connector; Claude Code: claude mcp add toolrouter -- npx -y toolrouter-mcp). Setup: https://toolrouter.com/docs/markdown/quickstart.
Hard Rules
- One URL, one product.
scrape_producttakes a direct product page, not a category or search page. Useweb-scraperfor anything else. - Use the exact operation. Pick the
toolandskillbelow, or calldiscover. Never invent an input field. - Prices are a snapshot. Report the currency and when you pulled it. Marketplace prices come from indexed listing text and can be stale; open the listing before relying on a price.
- Respect site terms. Marketplace search uses search-engine indexes, not logins. Do not try to bypass sign-in walls.
Authentication and Cost
The server provisions an anonymous account on first use. Paid calls need credits: if use_tool returns a claim link, give it to the user so they can sign in and add credits. discover shows the live price for each operation; listing_details and source_coverage are free.
Core Workflow
- Search the market.
marketplace-search/search_listingswith the item, region and condition. - Inspect listings.
marketplace-search/listing_detailson the URLs worth a closer look (free). - Scrape the stores.
product-scraper/scrape_producton competitor product pages for price, variants, specs and images. - Read the rest of the site.
web-scraper/scrape_pagefor pricing pages, shipping and returns policies or category pages. - Summarise. Table of source, title, price, currency, condition and URL, with the low, median and high price.
Essential Operations
| Job | tool | skill | Input |
| Full product details from a store page | product-scraper | scrape_product | {"url":"https://store.example.com/products/item","include_images":false} |
| Search marketplaces for an item | marketplace-search | search_listings | {"query":"sony wh-1000xm5","regions":["UK"],"condition":"used","limit":10} |
| One marketplace only | marketplace-search | search_listings | {"query":"herman miller aeron","source":"ebay","country":"gb"} |
| Preview one listing | marketplace-search | listing_details | {"url":"<listing url>"} |
| Compact price snapshot to track over time | marketplace-search | watchlist_snapshot | {"query":"steam deck oled","regions":["UK"]} |
| Which marketplaces cover a region | marketplace-search | source_coverage | {"regions":["EU"]} |
| Any other page as markdown | web-scraper | scrape_page | {"url":"https://example.com/pricing"} |
Call these through use_tool with tool, skill and input. Sources: shopping_index, ebay, facebook_marketplace, craigslist, gumtree, trademe, mercadolibre, olx, rakuten, vinted, carousell.
Common Patterns
- Competitor price check: scrape the same product on 3 to 5 competitor stores and line up price, variants and stock.
- Resale value:
search_listingswithcondition:"used"across eBay, Vinted and Facebook Marketplace; quote the price range, not one listing. - Sourcing: search by model number across regions, then open the cheapest listings with
listing_details. - Monitoring: run
watchlist_snapshoton a schedule and diff against the last snapshot the user saved; the tool does not store state. - Ad + product research: pair with
ad-library-scraperto see which products a competitor advertises, then scrape those product pages.
Gotchas
- Shopify stores are detected automatically. Other stores use AI extraction, so check odd fields against the page.
scrape_productsaves images to the user's file space by default; setinclude_images:falsewhen only data is needed.- Marketplace coverage depends on the search index for that market. Run
source_coveragebefore promising a country. - Results can include search or category pages and other regions. Keep rows whose
source_regionmatches the user's market and whose snippet names the exact item. - Listing prices may be missing or in local currency; never convert currencies without saying so.
Quick Reference
Connect → search listings → inspect → scrape product pages → price table with sources and dates.
More: https://toolrouter.com/tools/product-scraper · https://toolrouter.com/tools/marketplace-search · https://toolrouter.com/tools/web-scraper
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

