Building Agents with Amazon Nova Act and MCP
summary
Workshop shows building agents that browse and act on the web using Amazon Nova Act, MCP, Bedrock models and the open-source Strands framework. Hands-on demos run a coffee‑maker search and parallel monitor scrapes (multi‑threaded) from simple Python scripts, produce video logs and screenshots, and wire an MCP server to Bedrock models (Claude 3.5 sonnet) to generate docs, diagrams and a PowerPoint. Results: Nova Act reliably automates pages without captchas, captchas stop runs, a multi‑agent solution appears in under 50 lines of code, and a custom MCP server adapter ties Nova Act to other clients.
AI-written summary · about
chapters (13)
- 0:00Introduction to agents
- 1:13What is an agentic system
- 3:22Evaluating agentic systems
- 4:45Agents on AWS
- 6:10Strands agent framework
- 7:19Amazon Nova Act overview
- 8:22Modern Client Protocol
- 9:27Workshop setup
- 16:12Module 1 Nova Act demo
- 31:55Module 2 MCP server
- 43:03Module 3 Strands agents
- 53:16Multi-agent collaboration
- 1:17:10Q&A session
related talks
40mBuilding Agents with Agent Bricks and MCP
24mAgentic Al in SW Development: Evolving Patterns & Protocols
28mAgentic AI systems using Model Context Protocol
27mBuilding AI Agents with Spring & MCP
29mBuilding Interoperable Agents: When A2A meets MCP
49mBuilding AI Agents with Spring & MCP
36mPutting AI Agents to Work: A New Era of Open Source Connectivity with MCP
29mAnthropic's Claude in Amazon Bedrock