AgentQL MCP server

This is a small helper that lets your AI read a web page and pull out just the details you care about, like names, prices, or links. It is made by the team behind AgentQL, a service built for this exact job. It is handy if you often copy information from websites by hand into a spreadsheet or document.

What this MCP server does

You give your AI a web address and a plain description of the data you want from that page. Your AI passes both to this helper, which sends them to AgentQL. AgentQL opens the page, finds the matching pieces, and sends them back. Your AI then shows you the result, usually as a neat list or table in the chat.

Click to zoom

What you can do with it

• Pull a list of products with prices from a shopping page
• Collect names, job titles, and links from a team page
• Grab video titles, authors, and view counts from a search results page
• Turn a messy web page into a clean table you can paste into a document
• Extract contact details or addresses from a directory page
• Gather headlines and dates from a news or blog page

Try asking your AI

• “Extract the list of videos from https://www.youtube.com/results?search_query=agentql, with title, author, views, and url, and format it as a markdown table”
• “From this page, get every product name and its price, and skip the ads”
• “Pull the names and job titles of everyone listed on this team page”
• “Get the headlines and publish dates from this news page and put them in a table”

What it gives back to you

You get back the data you asked for, usually as a list or a table right in the chat. Each item has the fields you described, like a title, a price, or a link. If the page does not have something you asked for, that field may come back empty. You can then copy the result into a document or spreadsheet.

Before you start

Who it's for

Anyone who regularly copies information from websites by hand, like researchers, marketers, analysts, and assistants.