Skynet Monitor

An experimental literature piece disguised as an ncurses-style AI threat dashboard.

The app presents fictional AI risk data as a classified terminal instrument: threat boards, dossiers, field notes, transmission fragments, live logs, redactions, CRT effects, and multilingual interface text.

Design Canon

The look, feel, and writing voice are defined in DESIGN.md.

Future agents and contributors should also read AGENTS.md before changing UI, copy, data, or motion.

Tech Stack

• Next.js 14
• React 19
• TypeScript
• Tailwind CSS
• pnpm
• lightweight local i18n via i18n-react

Project Structure

app/
  page.tsx              Locale bootstrapping and top-level render
  globals.css           Tailwind setup plus CRT, redaction, and motion effects
components/
  skynet-dashboard.tsx  Main terminal dashboard experience
data/
  skynet-data.ts        Fictional threat data and localized content
lib/
  i18n.tsx              Locale provider, labels, and translation helpers

Getting Started

Install dependencies:

pnpm install

Run the local development server:

pnpm dev

Build for production:

pnpm build

Run TypeScript checks:

pnpm exec tsc --noEmit

Development Notes

• Keep the first screen as the artifact. Do not add a marketing landing page.
• Preserve English, Spanish, and French support for user-facing text.
• Keep the dashboard compact, terminal-native, and operational.
• Use motion sparingly: cursor blink, scanline, redaction, typewriter, subtle corruption.
• Treat copy as part of the interface, not explanatory decoration.

Deployment

The project is deployed from the repository through the existing GitHub/Vercel workflow.

The previous v0 project link is:

https://v0.app/chat/fHlFe78y9VL