agent-browser - Browser Automation CLI for AI Agents
Command-line interface for automating browser interactions including navigation, form filling, clicking, and data extraction
Tags
Updated: 2026-02-09Capabilities
Typical Inputs
Typical Outputs
What this skill does
- open websites
- fill forms
- click buttons
- take screenshots
- extract page text
- select dropdown options
- press keyboard keys
- scroll pages
- save browser state
- load browser state
- record sessions
- highlight elements
- capture PDF files
- tap mobile elements
Inputs
- URLs
- browser state files
- iOS simulators
- local file paths
- proxy configurations
- session identifiers
- element references
- CSS selectors
- semantic locators
Outputs
- screenshot files
- PDF documents
- session recordings
- page text files
- JSON data files
- state files
- terminal output
Requirements
- Bash environment
- agent-browser tool access
- macOS with Xcode (for iOS)
- Appium with xcuitest driver (for iOS)