cmux-browser Skill

End-user browser automation with cmux. Use when you need to open sites, inspect or interact with browser surfaces, wait for page state, and extract data without stealing focus.

Install
npx skills add 'https://github.com/manaflow-ai/cmux/tree/main/skills/cmux-browser'
Download bundle ↓
View on GitHub
← Back to SKILL.md

Video Recording

cmux browser exposes no built-in video recording command: automation runs on WKWebView, and the agent-browser recording pipeline is Chrome/CDP-specific. Related: commands.md, ../SKILL.md.

Capture evidence for flaky-automation debugging, CI logs, and release-to-release flow diffs with step screenshots and a snapshot timeline instead:

cmux browser --surface "$SURFACE" screenshot > /tmp/step1.b64
cmux browser --surface "$SURFACE" snapshot --interactive > /tmp/snap-1.txt
cmux --json browser --surface "$SURFACE" click e3 --snapshot-after > /tmp/action-1.json
cmux browser --surface "$SURFACE" screenshot > /tmp/step2.b64
cmux browser --surface "$SURFACE" snapshot --interactive > /tmp/snap-2.txt

Capture before and after each mutating action, add --snapshot-after on state-changing clicks/fills/types, and group artifacts by timestamp or run id. Use an external screen recorder when full-motion capture is genuinely required.

Referenced from SKILL.md