#569 · Primary category: AI Agents & Automation

openbrowser

ai-agents automation claude playwright puppeteer sandbox

Let AI agents browse the web. An autonomous toolkit for browser-based AI agents.

Project last updated:04/02/26

GitHub Stars

9.5K

Forks

865

Contributors

1

License

MIT

Why we included this project

This TypeScript framework gives AI agents a real browser to drive, so they can complete tasks on live websites instead of just answering questions. You describe what you want in plain language, and the agent handles the clicking, typing, scrolling, and data extraction itself, without you writing Playwright scripts by hand. It supports OpenAI, Anthropic, and Google models through the Vercel AI SDK, and switching providers is a single flag. The interactive REPL lets you step through a live session command by command, which makes debugging and prototyping much faster. There's also a sandboxed mode with CPU and memory limits and domain restrictions, so it's a workable option when you need to run agents in a constrained or untrusted environment.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category