Loading...
Loading...
Browser automation skill for AI agents using the mb CLI. Use when the agent needs to browse the web, take screenshots, scrape text, fill forms, click elements, record screencasts, run JS in pages, or audit designs. Triggers on: "browse", "open a page", "take a screenshot", "scrape", "fill form", "click button", "web automation", "record screen", "design audit", "accessibility check".
npx skill4agent add runablehq/mini-browser mini-browsermbmb# 1. Is mb installed?
which mb && echo "mb: ok" || echo "mb: MISSING"
# 2. Is Chrome listening on CDP?
curl -sf http://127.0.0.1:9222/json/version > /dev/null && echo "chrome: ok" || echo "chrome: NOT RUNNING"mbmbnpm install -g @runablehq/mini-browsermb-start-chrome--remote-debugging-port=9222mb-restart-chromemb go "https://example.com" && mb text| Variable | Default | Description |
|---|---|---|
| | CDP port |
| auto-detected | Path to Chrome/Chromium binary |
| | PID file location |
| | Chrome profile directory |
| Command | Description |
|---|---|
| Navigate to URL (waits for networkidle) |
| Print current URL |
| Go back |
| Go forward |
| Command | Description |
|---|---|
| Visible text content (default: body) |
| Screenshot to PNG (default: ./shot.png) |
| List interactive elements with coordinates |
| Command | Description |
|---|---|
| Click at coordinates |
| Type text (with coords: selects first) |
| Fill form fields by label/name/placeholder |
| Press keys (Enter, Tab, Meta+a) |
| Hover at coordinates |
| Drag between points |
| Scroll (default: down 500) |
| Command | Description |
|---|---|
| Start recording (.webm, .mp4, .gif) |
| Stop recording and save |
| Check if recording is active |
| Command | Description |
|---|---|
| List open tabs |
| Open new tab, print index |
| Close tab (default: last) |
| Command | Description |
|---|---|
| Run JavaScript in page context |
| Wait for ms / selector / networkidle / url:pattern |
| Design audit (palette, typography, contrast, a11y, SEO) |
| Stream console logs (Ctrl+C to stop) |
| Flag | Default | Description |
|---|---|---|
| 30000 | Command timeout |
| 0 | Target tab index |
| false | Structured JSON output |
| false | Right-click |
| false | Double-click |
| 30 | Recording frame rate |
| 1 | Recording scale factor |
mb snap # list interactive elements with (x, y)
mb click 512 380 # click the button at those coordinates
mb wait networkidle # wait for the page to settle
mb snap # observe againmb go "https://example.com/login"
mb fill "Email=user@example.com" "Password=hunter2"
mb key Enter
mb wait url:/dashboardmb shot page.pngmb text "main" # text from <main>
mb text "#content" # text from #content
mb text # full body textmb js 'document.title'
echo 'document.querySelectorAll("a").length' | mb js -mb record start demo.mp4 --fps 30 --scale 1
# ... interact with the page ...
mb record stopmb audit # human-readable report
mb audit --json # structured JSON outputmb js 'document.querySelector("[class*=cookie]")?.remove()'mb wait 2000 # sleep 2 seconds
mb wait ".modal" # wait for selector to appear
mb wait networkidle # wait for no network activity
mb wait url:/dashboard # wait for URL to contain stringsnaptexttext "main"text "p"gowait ".selector"typefill#.[--jsonsnap[{role, name, x, y, state}]tab list[{index, url, title}]logsaudit~/.mb-recorder.jsontab close| Problem | Fix |
|---|---|
| "Chrome not found" | Set |
| Connection refused | Run |
| Stale recording state | Delete |
| Chrome window wrong size | |
| Element not in snap output | |