⚠
High Risk:This skill has significant security concerns. Review the findings below before installing.

screen-monitor

⚠Caution·Scanned 2/17/2026

Provides WebRTC and browser-relay screen sharing for LLM-based visual analysis. It captures frames and POSTs them to /api/screen-frame, stores them at /tmp/clawdbot-screen-latest.png, exposes a LAN URL http://${LAN_IP}:18795/screen-share, and sets Access-Control-Allow-Origin: *.

from clawhub.ai·v5ca190f·13.1 KB·0 installs
Scanned from 1.0.1 at 5ca190f · Transparency log ↗
$ vett add clawhub.ai/emasoudy/screen-monitorReview security findings before installing

Screen-Monitor Skill

Browser-based screen sharing with vision analysis. Works with any multimodal LLM.

Installation

clawdhub install screen-monitor

Or manual:

git clone https://github.com/emasoudy/clawdbot-skills.git
cp -r clawdbot-skills/screen-monitor ~/.clawdbot/skills/

Usage

  1. Get share link: screen_share_link
  2. Open link in browser (e.g., http://192.168.1.100:18795/screen-share)
  3. Click "Start Sharing"
  4. Ask questions about your screen

Requirements

  • Multimodal LLM (Gemini, Claude, Qwen3-VL, etc.)
  • Modern browser with WebRTC
  • Optional: ImageMagick for OS fallback

License

MIT - See LICENSE file