BrowserSkill
Let AI agents use your real, logged-in browser without interrupting your work. CLI + extension for browser automation across any shell-capable AI agent.
BrowserSkill is an agent skill and automation toolkit that enables AI agents to control your already-logged-in Chrome or Edge browser without disrupting your workflow. It consists of a command-line daemon (bsk), a browser extension, and installable skills for Cursor, Claude Code, Codex, OpenClaw, and other shell-capable agents. Agent tasks run in a separate visible Agent Window while your normal browsing continues uninterrupted, with built-in human-in-the-loop support for captchas and confirmations. The system works locally via WebSocket and shell commands, supporting macOS, Linux, and Windows across x64 and ARM architectures.
Key Features
Use Cases
- 01Automating web research and data extraction on sites where you're already authenticated
- 02Testing web applications with real user sessions instead of separate test credentials
- 03Delegating repetitive browser tasks to AI agents while continuing other work uninterrupted
- 04Building agent workflows that require occasional human input for CAPTCHAs or multi-factor authentication
- 05Scraping or monitoring content from login-protected dashboards and social media accounts
- 06Creating browser automation scripts that agents can execute through natural language instructions
Related Skills
View moresuperpowers
An agentic skills framework & software development methodology that works.
skills
Skills for Real Engineers. Straight from my .agents directory.
skills
Public repository for Agent Skills
ponytail
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
BrowserSkill — FAQ
What is BrowserSkill and what does it do?+
BrowserSkill is an agent skill that connects AI assistants like Cursor, Claude Code, and Codex to your real Chrome or Edge browser. It lets agents automate browser tasks using your existing login sessions while you continue working in separate windows.
How do I install BrowserSkill?+
Install the bsk CLI via the provided shell script (macOS/Linux) or PowerShell script (Windows), then add the browser extension from the Chrome Web Store or Edge Add-ons. Finally, run 'bsk install-skill' and select your agent harness to install the skill definition.
Which AI agents and browsers work with BrowserSkill?+
BrowserSkill works with any shell-capable agent including Cursor, Claude Code, Codex, OpenClaw, CodeBuddy, WorkBuddy, Pi, Hermes Agent, and DeepSeek Harness. It supports Chrome and Microsoft Edge officially, with other Chromium browsers expected to work.
Does BrowserSkill require API keys or external services?+
No, BrowserSkill runs entirely locally on your machine. It uses WebSocket communication on 127.0.0.1 between the CLI daemon and browser extension, with no external API keys or cloud services required.
Is BrowserSkill free to use?+
Yes, BrowserSkill is open source and released under the MIT license. Both the CLI, browser extension, and DeepSeek Harness plugin are free to use.
Can BrowserSkill interrupt my normal browsing?+
No, agent browser tasks run in a separate Agent Window visible on screen, leaving your normal browser windows and tabs untouched unless the agent explicitly borrows a tab and returns it afterward.
How do I install BrowserSkill?+
Open the source repository on GitHub and follow its README. BrowserSkill is a skill — MCP Agents Market links you directly to the official repo.
Is BrowserSkill free?+
BrowserSkill is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.