Skip to content

installation.md

Latest commit

 

History

History
98 lines (64 loc) · 1.95 KB

installation.md

File metadata and controls

98 lines (64 loc) · 1.95 KB
Raw

Installation

Agent Integration (Recommended)

Tell your AI agent:

Install browser-act. Skill source: https://github.com/browser-act/skills/tree/main/browser-act . Verify it works after installation.

The agent will handle CLI installation, Skill configuration, and verify everything is working.

Manual Installation

uv tool install browser-act-cli --python 3.12

Verify:

browser-act --version

Authentication (Optional)

An API Key unlocks the following features:

  • stealth browsers — Anti-detection browsing with fingerprint spoofing
  • stealth-extract — One-command extraction of protected page content
  • Dynamic proxy — Managed IP rotation, allocated by region
  • solve-captcha — Automatic captcha solving

Chrome and chrome-direct browsers work without authentication.

Get an API Key:

browser-act auth login
# Opens registration link → complete signup → poll for the key
browser-act auth poll

Or set it directly:

browser-act auth set <your-api-key>

Upgrade

uv tool upgrade browser-act-cli

The Skill layer automatically detects mismatches between CLI and Skill content versions and guides the upgrade.

Platform Support

Platform Status
Windows Supported
macOS Supported
Linux Supported

Requirements

  • Python 3.12+
  • Chrome/Chromium (for chrome and chrome-direct types)
  • API Key (only required for stealth type)

Troubleshooting

Command not found after install

Make sure the uv tool directory is in your PATH:

uv tool dir
# Add the output path to your shell PATH

Diagnostics

Upload logs for issue diagnosis:

browser-act report-log

Send improvement suggestions:

browser-act feedback "Description of the issue or suggestion"

Next Steps

  • Quick Start — Run your first automation
  • Skills — How agents discover and use BrowserAct