Turn rough ideas into precise prompts—built for investment teams.

Meta Prompt Maker is a web application that helps users optimize their prompts through an AI-powered clarification process. Instead of struggling with prompt wording, the AI asks 5-6 targeted questions and returns a polished, copy-ready prompt that can optionally be run across multiple AI models (OpenAI GPT-4o, Anthropic Claude, Google Gemini) for comparison.

• Smart Prompt Optimization: AI analyzes your prompt and asks clarifying questions
• Use-Case Packs: Pre-built templates for investment industry scenarios
• Multi-Model Runner: Compare outputs across GPT-4o, Claude, and Gemini
• Analytics & Tracking: Anonymous usage analytics and performance metrics
• Rate Limiting: Built-in protection against abuse
• Responsive Design: Works on desktop and mobile

• Frontend: Next.js 14+ (App Router), TypeScript, Tailwind CSS, shadcn/ui
• Backend: Next.js API routes, Server Actions
• Database: PostgreSQL (Neon) with Prisma ORM
• AI Providers: OpenAI, Anthropic, Google Generative AI
• Validation: Zod
• Styling: Tailwind CSS with Inter font

1. Install Dependencies

   npm install

2. Environment Variables Create a .env file with:

   # Database
   DATABASE_URL="your-neon-database-url"
   
   # API Keys
   OPENAI_API_KEY="your-openai-key"
   ANTHROPIC_API_KEY="your-anthropic-key"
   GOOGLE_API_KEY="your-google-key"
   
   # Security
   RATE_LIMIT_SALT="your-random-salt"
   
   # Branding
   ALTITUDE_BRAND_COLOR="#0F2B5B"

3. Database Setup

   npx prisma migrate dev
   npx prisma generate

4. Start Development Server

   npm run dev

The application includes pre-built use-case packs for investment firms:

• LinkedIn Post: Professional social media content
• Investment Memo: Structured investment analysis
• RFP Response: Request for proposal responses
• Compliance Note: Regulatory communications
• Client Email: Professional client communications
• Portfolio Commentary: Investment performance commentary

• POST /api/optimize - Generate clarifying questions
• POST /api/finalize - Create optimized prompt
• POST /api/run - Run prompt across selected AI models
• POST /api/analytics - Log usage events

• General API: 60 requests/hour
• Optimization: 20 requests/hour
• Model runs: 10 requests/hour

• Session: User sessions with IP hashing
• PromptRecord: Complete prompt optimization records
• Event: Analytics and usage tracking

• IP address hashing for privacy
• Rate limiting by IP
• Input validation with Zod
• CORS protection
• Environment variable protection

1. Deploy to Vercel
2. Configure environment variables
3. Set up Neon database connection
4. Configure custom domain (e.g., meta.altitude7.com)

Altitude Global Advisors

Outputs may contain errors. Review before use. Nothing here is investment advice.