🤖 Autonomous multi-agent AI system for Etsy digital products
Scrape competitors → Analyze gaps → Generate PDFs → Track performance
4 agents, each with its own llm-council for multi-LLM deliberation:
- Scout — Scrapes Etsy shops, extracts listing data (Playwright)
- Analyst — Identifies market opportunities and gaps (pandas + council)
- Creator — Generates product content and PDFs (council + reportlab)
- Optimizer — Tracks performance and recommends improvements (council)
# Install dependencies
uv sync
uv run playwright install chromium
# Scrape a competitor shop
uv run etsy-store scrape "ShopName" -o data/shop.json
# Analyze market data
uv run etsy-store analyze data/shop.json
# Generate a digital product
uv run etsy-store generate "Daily Planner" -o outputetsy-store/
├── scout/ # Scraping agent (Playwright + BeautifulSoup)
├── analyst/ # Market analysis agent (pandas + llm-council)
├── creator/ # Content generation agent (llm-council + PDF)
├── optimizer/ # Performance tracking agent (analytics + council)
├── config/ # Configuration files (niches, pricing)
├── data/ # Scraped data and analytics
├── docs/ # Documentation
├── tests/ # Test suite
└── .claude/ # llm-council configs per agent
- Python 3.10+
- uv (package manager)
- Ollama (for local LLMs)
- Playwright browsers
MIT