LABSCO SUMMARY

One tool, and the query is a sentence instead of a selector you have to maintain.

extract-web-data takes a url and a prompt describing what you want and returns JSON, so the extraction logic lives in the request rather than in a scraper somebody has to keep fixing. There is no navigation, no click, no session: what a URL alone can render is what this can read.

What it is

A single-tool server over AgentQL's extraction service: a URL plus a natural-language description of the data, answered as JSON.

What you get

Structured JSON out of a page you name, with the shape described in prose rather than in selectors
A prompt-and-url call shape, so the same tool covers a new site without per-site configuration
Data rather than page source in the response, so a long page does not have to travel through the model's context to be read

Requirements

AGENTQL_API_KEY, and a URL the service can reach on its own. MIT.

Setup effort

One command plus a key — npm install -g agentql-mcp, then supply credentials

See installation options

USE CASES

01

Getting three fields off a page you do not want to parse

Give the URL and describe the data in a sentence — the fields come back as JSON, so nothing downstream has to read markup or maintain a selector for that site.

extract-web-data

1 tool

Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.

`extract-web-data`

Extracts structured data as JSON from a web page given a URL using a Natural Language description of the data.

BEFORE YOU INSTALL

What to know first

Tools you get

1 tool when we connected.

Requires

npx on PATH
AGENTQL_API_KEY in your environment

Where it runs

All 4 documented setups run the server on your machine.

CONTEXT COST

Light Standing cost

Rated by the size of the one tool definition that rides along on every turn

INSTALLATION

Connect your client

Server setup

TERMINAL

npm install -g agentql-mcp

MCP Clients

Claude Desktop, Windsurf

JSON

{
  "mcpServers": {
    "agentql": {
      "command": "npx",
      "args": [
        "-y",
        "agentql-mcp"
      ],
      "env": {
        "AGENTQL_API_KEY": "<YOUR_AGENTQL_API_KEY>"
      }
    }
  }
}

VS Code — user, workspace