agent-browser - Automate browser interactions from the command line
Automates website navigation, interaction, data extraction, screenshots, testing, and other browser tasks through a CLI.
Tags
Updated: 2026-10-02Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Navigate websites
- Inspect interactive elements
- Fill forms
- Click controls
- Extract page data
- Capture screenshots
- Save browser state
- Compare page states
- Automate iOS Safari
Inputs
- Website URLs
- Element references
- Form values
- Login credentials
- CSS selectors
- Browser state files
- Output file paths
- Baseline files
- iOS device names
Outputs
- Browser navigation state
- Element snapshots
- Extracted text or JSON
- Screenshots
- PDF files
- Video recordings
- Profile traces
- Saved authentication state
- Page difference results
Requirements
- agent-browser CLI
- Bash command access
- Chromium for rendered captures
- macOS with Xcode for iOS automation
- Appium with XCUITest driver for iOS automation