Implementing the AgentQL Mannequin Context Protocol (MCP) server

agentql Scrub your web site with unstructured information by defining the precise form of the knowledge you want. Even from dynamic content material and continuously altering layouts, you may get constant and structured outcomes.

On this tutorial, you’ll implement an AgentQL MCP server throughout the Claude desktop and use Claude’s built-in visualization capabilities to discover your information. Particularly, rub the search outcomes web page on Amazon ai booksextract particulars resembling worth, score, variety of critiques, and so on.

Step 1: Setting Dependencies

Node JS

NPX is required to run the AgentQL server that comes with node.js.

• Obtain the most recent model of node.js nodejs.org
• Run the installer.
• Go away all settings as default and full set up

Claude Desktop

Obtain utilizing Claude https://claude.ai/download.

agentQl API

Create an agentQL API key dev.agentql.com/api-keys Please retailer it safely. You will want it later on this tutorial.

Step 2: Putting in the bundle

As soon as node.js is put in, open the terminal and run the next command:

npm set up -g agentql-mcp

Step 3: Configuring the MCP Server

Subsequent, configure Claude to connect with the MCP server. It’s going to open claude_desktop_config.json Information within the Claude set up listing utilizing any textual content editor. If the file doesn’t exist, you may create it manually. As soon as opened, enter the next code:

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

exchange with the generated key.

Step 4: Run the server

As soon as the MCP configuration is full, the server ought to seem in Claude. AgentQL Server features a single highly effective device (Extract_Web_Data) that retrieves the URL and pure language description of the information construction to be extracted.

You should utilize any URL to scrape. On this tutorial, we requested Claude to visualise the extracted information utilizing the Amazon Search Outcomes web page in AI Books. Claude gives an interactive terminal that generates code that processes and visualizes information. You possibly can edit the code if vital. As soon as the code is confirmed, Claude presents a bar chart with interactive choices to discover worth, scores, overview counts, and even worth and score scatter plots, together with key abstract statistics.

You should utilize AgentQL to shave your web site, hook up with ideas and different servers resembling GitHub to routinely ship structured information for doc, monitoring, or additional automation.

This makes AgentQL a strong device for turning unstructured net content material into actionable insights. That is all easy and inside a pure language workflow.

Here is a fast overview of what is constructed with MarkTechPost:

I’m a civil engineering graduate (2022) from Jamia Milia Islamia, New Delhi, and have a powerful curiosity in information science, significantly neural networks and purposes in a wide range of fields.