# Copy to .env (gitignored) and fill in.
# Any OpenAI-compatible /chat/completions endpoint works.
LLM_BASE_URL=https://api.example.com/v1
LLM_API_KEY=your-key-here
LLM_MODEL=your-model-id
REPORT_SALT=replace-with-a-long-random-secret
REQUIRE_AUTH=0