Scout

Your personal deal-hunting research agent. scout.grove.place

Scout is an async shopping research tool that eliminates the cognitive overload of deal-hunting. Tell it what you want, walk away, and come back to a clean list of 5 perfect matches.

The Pitch

Shopping sucks when you have ADHD. Amazon is overwhelming. Deal sites are chaos. Black Friday is a nightmare.

Scout fixes this:

1. Tell it what you want — "cozy sweater, earth tones, under $60"
2. Walk away — Scout's AI agents search the web in the background
3. Come back to results — 5 curated products with direct purchase links

No infinite scrolling. No decision paralysis. No accidental $200 Amazon cart.

Current Status

Scout is a working MVP with core functionality implemented:

• Authentication: Google & Apple OAuth with session management
• Search Pipeline: AI‑powered multi‑agent search (Brave Search + Claude)
• Results Curation: Returns 5 curated products with shareable links
• Basic Billing: Credit‑based system with Stripe integration
• Profile & Dashboard: User preferences and search history

The system is deployed at scout.grove.place and ready for beta testing.

Tech Stack

Layer	Technology
Frontend	SvelteKit
Backend	Cloudflare Workers
Database	Cloudflare D1 (SQLite)
Queue	Cloudflare Queues
Cache	Cloudflare KV
AI	Claude API (Anthropic)
Search	Brave Search API
Payments	Stripe
Email	Resend

Development

Prerequisites

• Node.js 20+
• Wrangler CLI (npm install -g wrangler)
• Cloudflare account with D1, KV, and Queues access

Setup

# Install dependencies
npm install

# Set up Cloudflare resources (run these once)
npx wrangler d1 create scout-db
npx wrangler kv:namespace create SCOUT_CACHE
npx wrangler queues create scout-search-jobs

# Update wrangler.toml with the returned IDs

# Run initial database migration
npm run db:migrate

# Set secrets (for production)
npx wrangler secret put ANTHROPIC_API_KEY
npx wrangler secret put BRAVE_API_KEY
npx wrangler secret put GOOGLE_CLIENT_ID
npx wrangler secret put GOOGLE_CLIENT_SECRET
# ... etc

# Start development server
npm run dev

Scripts

• npm run dev - Start Vite dev server
• npm run dev:wrangler - Start with Cloudflare bindings
• npm run build - Build for production
• npm run preview - Preview production build locally
• npm run deploy - Deploy to Cloudflare Pages
• npm run check - TypeScript type checking
• npm run db:migrate - Run database migrations

Project Structure

scout/
├── src/
│   ├── lib/
│   │   ├── server/      # Server-only code (db, auth)
│   │   └── types.ts     # Shared types
│   ├── routes/
│   │   ├── api/         # API endpoints
│   │   ├── auth/        # Auth pages
│   │   ├── dashboard/   # User dashboard
│   │   ├── search/      # Search pages
│   │   └── s/[token]/   # Shareable results
│   ├── app.d.ts         # App types
│   └── hooks.server.ts  # Request hooks
├── docs/                # Project specs
├── migrations/          # D1 migrations
└── wrangler.toml        # Cloudflare config

Documentation

• docs/PROJECT_SPEC.md - Full product specification
• docs/DATA_MODEL.md - Database schema
• docs/AGENT_PROMPTS.md - AI agent prompt engineering
• docs/TODO.md - Roadmap and milestones

Internal Codename

The project is internally called Research Goblin. Feel free to use this in code comments, logs, and internal docs. Customer-facing name is Scout.

License

Proprietary. This is Autumn's product idea.