Space procurement intelligence demo built with Next.js for Vercel deployment.

• Natural language capability search
• Document upload for text or PDF inputs
• OpenAI-based extraction of three search angles:
  • Buyers
  • Programs
  • Signals
• Side-by-side neural search and Google comparison results
• Domain filtering for space and aerospace sites
• Search request payload inspection in the configuration panel
• Result sorting by relevance or recency
• Category filters and graceful API error handling

1. Install dependencies:

npm install

2. Create .env.local from .env.example and add your API keys.

3. Start the app:

npm run dev

• EXA_API_KEY: Required for the neural search provider
• OPENAI_API_KEY: Required for capability-angle extraction
• OPENAI_MODEL: Optional override for the OpenAI model
• GOOGLE_API_KEY: Optional, enables Google comparison search
• GOOGLE_CSE_ID: Optional, required with GOOGLE_API_KEY for Google comparison search

1. Import the repo into Vercel.
2. Set the environment variables from .env.example.
3. Deploy with the default Next.js settings.