# Copy to .env and fill in your OpenAI key:  cp .env.example .env
# The demos load .env automatically (see shared/llm.py).

# OpenAI credentials — the provider every demo uses by default.
OPENAI_API_KEY=<your-openai-key>

# Optional overrides:
# OPENAI_MODEL=gpt-4o

# Local fallback (no network) — uncomment to use Ollama instead of OpenAI:
# LLM_PROVIDER=ollama
# OLLAMA_MODEL=llama3.2
# OLLAMA_BASE_URL=http://localhost:11434