← ClaudeAtlas

agent-browserlisted

Headless browser automation using agent-browser CLI. This skill should be used when performing headless/background web scraping, parallel multi-site operations, or simple page queries that don't require login. Triggers on requests like "크롤링", "스크래핑", "페이지 정보", "헤드리스", "병렬로 웹 작업", "여러 사이트 동시에".
SONGYEONGSIN/vibe-flow · ★ 1 · AI & Automation · score 74
Install: claude install-skill SONGYEONGSIN/vibe-flow
# Agent Browser ## Overview Headless browser automation tool for AI agents. Use agent-browser CLI instead of Claude in Chrome when: - Headless/background operation is needed (no visible browser) - Parallel multi-site scraping or monitoring - Simple page queries without login requirements - Batch automation tasks ## Prerequisites Ensure agent-browser is installed before use: ```bash # Check installation which agent-browser || npm install -g agent-browser # Install Chromium if not present agent-browser install ``` ## Quick Start ### Basic Page Access ```bash # Open a page agent-browser open https://example.com # Get page snapshot (accessibility tree - AI optimized) agent-browser snapshot # Interactive elements only (compact) agent-browser snapshot -i # Screenshot agent-browser screenshot ./capture.png ``` ### Element Interaction ```bash # Click by selector agent-browser click "#submit-button" # Click by text (semantic) agent-browser find text "Login" click # Fill input field agent-browser fill "#email" "test@example.com" # Get text content agent-browser get text ".price" ``` ## Parallel Operations Use sessions for independent browser instances: ```bash # Create multiple sessions agent-browser --session site1 open https://site-a.com agent-browser --session site2 open https://site-b.com agent-browser --session site3 open https://site-c.com # Work in each session agent-browser --session site1 snapshot -i agent-browser --session site2 get text ".product-price"