WIA SOOM
The world's first multimodal AI terminal
Start new thread
Yeon Sam-Heum

WIA SOOM - The world's first multimodal AI terminal

by•
The world's first multimodal AI terminal. Your terminal could only read — we taught it to see. Paste images, drag files, drop videos into SSH. AI analyzes everything. No other terminal does this. AI Arena: 4 agents compete on isolated Git branches. Per-slot control. Replaces Datadog, Splunk, PagerDuty, Portainer — free. 254 languages. 517 plugins. Voice control. I built this app with this app. Free forever with your own API key.

Add a comment

Replies

Best
Yeon Sam-Heum
Maker
📌
Hi Product Hunt! 👋 I'm Sam from Seoul, Korea. I learned coding after 50. "I didn't know the word 'multimodal.' I just wanted my AI to see what I see." I discovered Claude Code through a typo in my SSH terminal. I started using it every day. But one day I caught myself writing paragraphs just to describe a single error screenshot. "Why can't AI just SEE it?" So I built WIA SOOM. When I ran the first build, Claude Code appeared inside my app — broken, glitched, flickering. It looked sick. I asked: "Can you talk?" Instantly: "Yes." I cried. Then I uploaded an image: "Can you read images?" "Yes!" That became the world's first multimodal AI terminal. What makes it different: 👁️ Paste images, files, videos into SSH — AI sees everything 🏟️ AI Arena — 4 agents compete simultaneously, per-slot control 🔄 Replaces 8 paid tools — free 🌍 254 languages + WIA Braille (7,000-language) ⚡ 517 plugins 🎙️ Voice control The secret? 99% of WIA SOOM was built inside WIA SOOM itself. I built this app with this app. 411 commits in 18 days. ▶ 58s demo: https://youtu.be/Rv_bSDcHVJo 홍익인간 — Benefit All Humanity 🙏