agent-browser - Browser automation tool for web testing and interaction
Automates browser interactions including navigation, element interaction, form filling, screenshots, and data extraction.
Tags
Updated: 2026-09-20Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Navigate web page URLs
- Capture interactive page DOM snapshots
- Interact with page elements
- Take page screenshots and PDFs
- Record browser session videos
- Manage cookies and local storage
- Intercept network requests and responses
- Execute JavaScript code on page
Inputs
- Target web URLs
- Element selectors and references
- Form input data
- Session state files
- Output file paths
Outputs
- DOM snapshot data
- Extracted text and attributes
- Screenshot and PDF files
- Session video files
- Saved state authentication files
- Console logs and error messages
Requirements
- agent-browser CLI command or Bun
- Network connection to target web pages