ScrAPI MCP

How to Install and Use ScrAPI MCP

ScrAPI MCP is a tool that helps you scrape web pages. It is easy to install and use. This guide will show you step-by-step how to set it up and use it with example code.

Step 1: Get an API Key (Optional)

Before you start, you can get an API key from the ScrAPI website. Having an API key is optional but recommended. Without it, you will only be able to make one call at a time and get 20 free calls a day. The key helps you use the service better.

Step 2: Choose Your Running Method

You can run ScrAPI MCP in two main ways: using Docker or NPX. Docker lets you run the server in a container. NPX runs it directly with Node.js. Both methods will work fine.

---

Step 3: How to Run ScrAPI MCP with Docker

If you like Docker, here’s what to do. First, make sure you have Docker installed.

Then add this to your claude_desktop_config.json file:

{
  "mcpServers": {
    "ScrAPI": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "SCRAPI_API_KEY",
        "deventerprisesoftware/scrapi-mcp"
      ],
      "env": {
        "SCRAPI_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Replace <YOUR_API_KEY> with your actual API key. If you don’t have one, you can leave it blank or skip the variable.

Step 4: How to Run ScrAPI MCP with NPX

If you want to run ScrAPI MCP with NPX (no Docker needed), use this example code in your config:

{
  "mcpServers": {
    "ScrAPI": {
      "command": "npx",
      "args": [
        "-y",
        "@deventerprisesoftware/scrapi-mcp"
      ],
      "env": {
        "SCRAPI_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Again, replace <YOUR_API_KEY> with your key.

---

Step 5: Run the Server Yourself (Optional Build)

If you want to build your own Docker image, use this command:

docker build -t deventerprisesoftware/scrapi-mcp -f Dockerfile .

This command creates the Docker image of ScrAPI MCP on your machine.

---

Step 6: Use ScrAPI MCP Tools

After setting up, you can use two main tools to scrape web pages:

1. scrape_url_html
   Use a URL and get the website content as HTML.
   Input: a string URL
   Output: HTML content

2. scrape_url_markdown
   Use a URL and get the website content as Markdown.
   Input: a string URL
   Output: Markdown content

These tools are helpful if a website blocks scraping or has geolocation restrictions.

---

Summary

ScrAPI MCP is a simple server to scrape web pages. You can run it with Docker or NPX. Get an API key if you want extra calls and better limits. Use the scrape_url_html or scrape_url_markdown tools to get website data easily.

Follow these steps, and you’ll be scraping websites in no time!