BrowserSkill is a local browser-automation bridge that lets AI agents work inside a user’s existing logged-in Chromium browser. It connects tools such as Claude Code, Codex, Cursor, Pi, Hermes Agent, and other shell-capable agents through the bsk CLI and a browser extension. Automated tasks run in a separate visible Agent Window so normal browsing can continue independently. Existing user tabs can be borrowed only through an explicit handoff workflow and returned afterward. Human intervention is supported for CAPTCHA, login, confirmation, and other steps that require manual action. The system can navigate pages, fill forms, capture screenshots, inspect content, and automate authenticated workflows. It supports macOS, Linux, and Windows with Chrome and Edge as primary browsers.

Features

  • Reuse of existing browser login sessions
  • Separate visible Agent Window for automation
  • Support for many shell-capable AI agents
  • Explicit borrowing of existing browser tabs
  • Human-in-the-loop handoff for blocked steps
  • CLI, daemon, and Chromium extension architecture

Project Samples

Project Activity

See All Activity >

Categories

AI Agents

License

MIT License

Follow BrowserSkill

BrowserSkill Web Site

Other Useful Business Software
Build Agents and Models on One Platform Icon
Build Agents and Models on One Platform

Everything you need to build production-ready agents and models. Access 200+ Google and third-party AI models and tools.

Gemini Enterprise Agent Platform is Google Cloud's comprehensive platform for developers to build, scale, govern, and optimize agents and models. Choose from Google's most advanced models and third-party models like Anthropic's Claude Model Family.
Start Free
Rate This Project
Login To Rate This Project

User Reviews

Be the first to post a review of BrowserSkill!

Additional Project Details

Operating Systems

Linux, Mac, Windows

Programming Language

TypeScript

Related Categories

TypeScript AI Agents

Registered

2026-09-17