Fetch Web Pages and Extract Data Through ScrapingBee and Your Muse
Through a custom connector, your muse can fetch web pages through ScrapingBee in plain words: read a page as text, pull out the fields you name, take a screenshot, and run Google and Amazon searches that come back sorted.
What it does
The connector adds 8 tools. One reads a page and returns its HTML, clean text, or a readable text version, using rotating, premium, or stealth proxies when a page blocks normal visits. One extracts data instead of the page, returning a neat list built from CSS or XPath rules, or from plain-language rules based on the fields you name. One takes a screenshot of the view, the full page, or a single element, on desktop or mobile. Two run Google searches: a quick result list, or a full parsed page with ads, map results, and related questions. Two search Amazon: products for a search term on a marketplace you pick, and one product read in full by its ASIN, with price, buy box, reviews, and stock. The last tool reports the credits used and left on your plan, the only one that costs nothing.
How to use it
- Start an AnythingMCP account and add the ScrapingBee connector.
- Copy your API key from the ScrapingBee dashboard account page.
- Paste the key into the connector settings as SCRAPINGBEE_API_KEY.
- In Muse, open Settings, then Connectors, and add a custom connector with the address AnythingMCP gives you. Sign in and approve.
- Ask for the prices on a page as a list, a screenshot of a page on a phone, or the top Amazon results for a product.
What you need
- A Muse account that can add custom connectors
- An AnythingMCP account with the ScrapingBee connector added
- A ScrapingBee account with an API key
What to watch for
- Every tool except the usage check spends plan credits, and the cost jumps with harder settings: a plain page costs 1 credit, a page with JavaScript costs 5, and a stealth proxy page costs 75. The muse asks before a batch of calls, and the free trial starts with 1,000 credits.
- Answers are capped at about 100 KB, so for long pages ask for text or named fields instead of the whole page.
- The connector reads pages. It does not send forms, forward posts, download files over 2 MB, or cover the Walmart and YouTube endpoints.
- This is a custom connector, and Meta does not review custom connectors. You can disconnect it in Settings at any time. The muse prepares, you decide.