| name | Agent Browser |
| description | Fast headless browser automation for AI agents (Rust-based with Node.js fallback). Navigate, click, type, and snapshot pages via structured commands. Triggers on browser automation, headless browsing, page snapshot. NOT for: websites needing login/cookies (use bb-browser or browser profile), or simple URL content extraction (use web_fetch/firecrawl). |
| read_when | ["Automating web interactions","Extracting structured data from pages","Filling forms programmatically","Testing web UIs"] |
| metadata | {"clawdbot":{"emoji":"🌐","requires":{"bins":["node","npm"]}}} |
| allowed-tools | Bash(agent-browser:*) |
Browser Automation with agent-browser
Fast headless browser automation (Rust-based). Navigate, click, type, and snapshot pages.
Quick Start
agent-browser open <url>
agent-browser snapshot -i
agent-browser click @e1
agent-browser fill @e2 "text"
agent-browser close
Core Workflow
- Navigate:
agent-browser open <url>
- Snapshot:
agent-browser snapshot -i (elements with refs like @e1, @e2)
- Interact using refs from the snapshot
- Re-snapshot after navigation or significant DOM changes
Key Notes
- Refs (
@e1, @e2) are stable per page load but change on navigation
- Always snapshot after navigation to get new refs
- Use
fill instead of type for input fields (ensures text is cleared)
- On Linux ARM64, use the full path in the bin folder
Route Table