This enhancement allows automation scripts and AI agents to perform actions not accessible through Chrome DevTools Protocol alone, enabling automation of native system dialogs, print handling, complex UI workflows, and capturing full-desktop screenshots. It broadens the range of web workflow automation scenarios including test automation and AI-driven web interactions.
The addition supports mouse movements, clicks, drags, scrolls and keyboard operations like typing and shortcuts at OS-level coordinates, improving the ability to interact with the entire browser environment. Available by default across all regions supporting AgentCore Browser, it benefits developers building more complex browser automation via LLM-powered tools and test frameworks.
Source & further reading
Amazon Bedrock AgentCore Browser adds OS-level interaction capabilities ↗Source published 2026-04-08. This report was prepared with AI assistance from the linked primary-source material. Claims and limitations should be checked against the original announcement or research.