Pix3lPrompt

Pix3lPrompt is an intelligent prompt editor for AI image, video, and audio generators. Build and optimize prompts for Midjourney, Flux, Stable Diffusion, DALL-E, Kling, Runway, Suno, Udio, and more — directly in your browser. All data is stored locally in IndexedDB: no backend, no account required.

Repository: github.com/Pix3ltools-lab/pix3lprompt

Tech Stack

• Framework: Next.js (App Router)
• Language: TypeScript
• Styling: Tailwind CSS 4 + shadcn/ui
• State Management: Zustand
• Storage: IndexedDB via Dexie.js (100% client-side)
• Icons: Lucide React
• Theme: next-themes (dark/light mode)
• Analytics: Vercel Analytics

Features

Multi-model Prompt Builder

Model-aware prompt assembly with the correct syntax for each generator:

• Midjourney: --ar, --v, --no parameters and parameter-aware formatting
• Flux / Stable Diffusion: Natural language prompts
• DALL-E: OpenAI-specific formatting
• Suno / Udio: Audio prompt syntax for music and sound generation
• Kling / Runway: Video generation prompt formatting

Visual Editor

• Subject field as the primary input
• 20 style chips (cinematic, photorealistic, watercolor, etc.)
• 14 lighting presets
• Aspect ratio picker
• Negative prompt field
• Color-coded live preview — each segment type highlighted differently

AI-Powered Optimization

Connect to an AI provider to refine and improve your prompts:

• OpenRouter (multi-model)
• OpenAI
• Anthropic
• LM Studio (local models)

Provider configuration is handled in the Settings page — no environment variables required.

Prompt History

• Save prompts with metadata: model, style, rating (1–5 stars), tags, favorites
• Search and filter saved prompts
• Reload any saved prompt into the editor
• Quick-start templates

Progressive Web App (PWA)

• Installable on desktop and mobile as a standalone app
• Offline support — works without an internet connection

Layout

• Desktop: 3-column layout (History | Editor | Templates)
• Mobile: Tab-based layout with bottom navigation

Environment Variables

No environment variables are required for basic usage. AI provider credentials (API keys) are entered directly in the app’s Settings page and stored locally in the browser.

Running Locally

git clone https://github.com/Pix3ltools-lab/pix3lprompt.git
cd pix3lprompt

npm install
npm run dev    # http://localhost:3000

Deployment

Vercel

1. Push to GitHub
2. Import the repository in Vercel
3. Set Framework Preset to Next.js
4. Deploy — no environment variables needed

Docker

git clone https://github.com/Pix3ltools-lab/pix3ltools-deploy.git
cd pix3ltools-deploy
./setup.sh
# Pix3lPrompt runs on http://localhost:3002

The Docker image uses a multi-stage build (node:20-alpine), runs as non-root user (uid 1000), exposes port 3002. No database or storage volumes required.

npm Scripts

npm run dev    # Development server
npm run build  # Production build
npm run start  # Start production server
npm run lint   # ESLint