agent-browser

1# Browser Automation with agent-browser
2 
3## Core Workflow
4 
5Every browser automation follows this pattern:
6 
71. **Navigate**: `agent-browser open <url>`
82. **Snapshot**: `agent-browser snapshot -i` (get element refs like `@e1`, `@e2`)
93. **Interact**: Use refs to click, fill, select
104. **Re-snapshot**: After navigation or DOM changes, get fresh refs
11 
12```bash
13agent-browser open https://example.com/form
14agent-browser snapshot -i
15# Output: @e1 [input type="email"], @e2 [input type="password"], @e3 [button] "Submit"
16 
17agent-browser fill @e1 "user@example.com"
18agent-browser fill @e2 "password123"
19agent-browser click @e3
20agent-browser wait --load networkidle
21agent-browser snapshot -i  # Check result
22```
23 
24## Essential Commands
25 
26```bash
27# Navigation
28agent-browser open <url>              # Navigate (aliases: goto, navigate)
29agent-browser close                   # Close browser
30 
31# Snapshot
32agent-browser snapshot -i             # Interactive elements with refs (recommended)
33agent-browser snapshot -i -C          # Include cursor-interactive elements (divs with onclick, cursor:pointer)
34agent-browser snapshot -s "#selector" # Scope to CSS selector
35 
36# Interaction (use @refs from snapshot)
37agent-browser click @e1               # Click element
38agent-browser fill @e2 "text"         # Clear and type text
39agent-browser type @e2 "text"         # Type without clearing
40agent-browser select @e1 "option"     # Select dropdown option
41agent-browser check @e1               # Check checkbox
42…