MCP Agents

@CodeWithHarshAI

Basic information

About MCP Agents

MCP Agents is an AI-powered browser automation tool that lets you interact with websites using natural language. Built with Streamlit, OpenAI, and Puppeteer via the Model Context Protocol (MCP), it supports multi-step navigation, interaction, and content extraction—all with simpl

#automation #browser #mcp #openai #streamlit #mcp-server #mcp-client

MCP Agents config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "MCP-Agents": {
      "command": "node",
      "args": [
        "--version",
        "&&",
        "npm",
        "--version"
      ]
    }
  }
}

MCP Agents tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is MCP Agents?

MCP Agents is an AI-powered interactive browser assistant built with Streamlit, OpenAI, and Puppeteer using the Model Context Protocol (MCP). It allows users to type natural language commands to navigate, click, scroll, and extract content from websites hands‑free.

How to use MCP Agents?

Install Python 3.8+, Node.js + npm, and an OpenAI API key. Clone the repository, install Python dependencies, optionally start the Puppeteer agent with npx -y @modelcontextprotocol/server-puppeteer, then run streamlit run main.py. Create a mcp_agent.secrets.yaml file with your OpenAI API key to keep credentials secure.

Key features of MCP Agents

Interact with websites using simple English commands.
Visual automation with screenshots, clicks, and scrolling.
Flexible agent system powered by the modular MCP framework.
Secure API key storage with optional secrets config.
Fully interactive UI showing command results.

Use cases of MCP Agents

Product scraping, news summarization, and automated browsing workflows.
Educational bots that walk through websites.
Daily content extraction from dynamic pages.

Frequently asked questions

What are the requirements to run MCP Agents?

Python 3.8 or newer, Node.js with npm, and an OpenAI API key.

How do I set up secrets?

Create a file named `mcp_agent.secrets.yaml` with your OpenAI API key and ensure this file is listed in `.gitignore`.

What kind of commands can I give?

Examples include "Go to www.wikipedia.org", "Search for black holes and summarize the page", "Click on the first heading", "Take a screenshot of the hero section", and "Scroll down and extract all h2 titles".

How do I start the application?

After installing dependencies and optionally starting the Puppeteer agent, run `streamlit run main.py` and visit `http://localhost:8501` in your browser.

What license is MCP Agents under?

MCP Agents is licensed under the MIT License.

Comments

More Browser Automation MCP servers

Browse Bridge

Share selected Chrome tabs with AI agents. You enable the tab you want to share and maintain the ability to instantly take control. Browse Bridge is a remote MCP server available at https://mcp.browsebridge.com. What c