Pull Clean Data From Web Pages Through Diffbot and Your Muse

Through a custom connector, your muse can turn messy web pages into clean data in plain words: the price from a product page, or the author and date from an article.

What it does

The connector adds 11 tools, and every one only reads. One tool checks the account, showing the plan, monthly credits, and daily use, at no cost. One tool reads any page, works out what kind of page it is, and pulls it apart the right way. Separate tools pull an article, a product, a discussion, and an image or video page, each with the details that matter for that kind of page. One tool searches a large index of companies, people, articles, products, and events, and one matches a single company or person in that index from details you have. One reads text you paste and finds the names, companies, and feeling in it, and one searches the web index and returns ranked pages with excerpts.

How to use it

1. Start an AnythingMCP account and add the Diffbot connector.
2. Create a Diffbot account, open the dashboard, and copy your API token. Enter it only where the connector asks.
3. In Muse, open Settings, then Connectors, and add a custom connector with the address AnythingMCP gives you. Sign in and approve.
4. Ask the muse to check the account first, so you see your plan and credits.
5. Give it one page you know well, such as a product page, and check the price and stock state against the page itself.
6. Before any index search, ask what it will cost, and keep the result list small.

What you need

• A Muse account that can add custom connectors
• An AnythingMCP account with the Diffbot connector added
• A Diffbot account and API token

What to watch for

• Almost every tool spends credits. A page pull costs 1 credit, or 2 through the backup route. Index results cost 25 credits for each company or person returned, so a wide search gets expensive fast.
• Searches that find nothing are free, and the account check is free. Ask for the cost first on any search that could return many results.
• A page behind a login, or one that blocks outside readers, may come back empty. That is a limit of the page, not a sign the connector is broken.
• 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.