sample-agentic-ai-web
by aws-samples·★ 50·Score 45
Amazon Bedrock web automation assistant with MCP server for browser automation and vision capabilities.
Overview
This repository provides a comprehensive implementation of an MCP server for web automation using Amazon Bedrock with Anthropic Claude and Amazon Nova models. It demonstrates a progressive approach to building sophisticated web automation capabilities, from basic API calls to advanced features like vision-based element interaction, form submission, and conversation history management. The implementation leverages FastMCP framework for structured tool definitions and proper resource lifecycle management.
Try asking AI
After installing, here are 3 things you can ask your AI assistant:
When to choose this
Choose this MCP server if you need a production-ready web automation solution with vision capabilities and are already using AWS Bedrock.
When NOT to choose this
Don't choose this if you need vendor-agnostic solutions or aren't using AWS Bedrock, as it has tight AWS integration.
Tools this server exposes
7 tools extracted from the READMEnavigateNavigate to a specified URL using Playwright
screenshotCapture screenshots of web pages
ask_humanAsk the user questions during execution
clickClick on elements based on visual analysis
typeType text into previously clicked elements
scrollScroll up or down to see more content on long pages
write_fileSave search results and findings to markdown files
Comparable tools
Installation
# Clone the repository
git clone https://github.com/aws-samples/sample-agentic-ai-web.git
cd sample-agentic-ai-web
# Install dependencies
pip install -r requirements.txt
# Run the MCP server
python 10-mcp-server.pyFor Claude Desktop configuration, add to claude_desktop_config.json:
{
"mcpServers": {
"bedrock-web-tools": {
"command": "python",
"args": ["path/to/10-mcp-server.py"]
}
}
}Compare sample-agentic-ai-web with
Last updated · Auto-generated from public README + GitHub signals.