
If you're automating browsers through Claude's CLI and want something simpler than Playwright, this wraps Vercel's headless browser tool with a snapshot and ref pattern that works well for AI. You navigate to a page, run snapshot to get interactive elements tagged as @e1, @e2, etc., then use those refs to click, fill forms, or scrape. It handles the common stuff like screenshots, PDFs, and waiting for elements. The workflow is straightforward: snapshot after each navigation to get fresh refs, then act on them. Good for bash-based automation where you want readable commands instead of diving into full browser testing frameworks.
$ npx -y skills add eyadsibai/ltk --skill agent-browser --agent claude-codeInstalls into .claude/skills of the current project.
Vercel's headless browser automation CLI designed for AI agents. Uses ref-based selection (@e1, @e2) from accessibility snapshots.
# Check installation
command -v agent-browser >/dev/null 2>&1 && echo "Installed" || echo "NOT INSTALLED"
# Install if needed
npm install -g agent-browser
agent-browser install # Downloads Chromium
The snapshot + ref pattern is optimal for LLMs:
agent-browser open https://example.com
agent-browser snapshot -i # Get refs
agent-browser click @e1 # Use ref
agent-browser fill @e2 "text"
agent-browser snapshot -i # Re-snapshot
agent-browser open <url> # Navigate to URL
agent-browser back # Go back
agent-browser forward # Go forward
agent-browser reload # Reload page
agent-browser close # Close browser
Sponsors