.env.example

# ─────────────────────────────────────────────────────────────
# Immich connection settings (REQUIRED — ask the user for both)
# ─────────────────────────────────────────────────────────────
# API_URL: the user's Immich server address. The value below is the
# typical localhost default; confirm the user's actual deployment URL.
IMMICH_API_URL=http://localhost:2283
# API_KEY: create one in the Immich Web UI:
#   Account Settings → API Keys → New API Key
IMMICH_API_KEY=your-immich-api-key

# ─────────────────────────────────────────────────────────────
# VLM settings (URL required; MODEL_NAME can be auto-discovered)
# ─────────────────────────────────────────────────────────────
# Any OpenAI-compatible endpoint works. Common examples:
#   llama.cpp:  http://localhost:8080/v1
#   vLLM:       http://localhost:8000/v1
#   Ollama:     http://localhost:11434/v1
#   LM Studio:  http://localhost:1234/v1
#   OpenAI:     https://api.openai.com/v1
#
# If the user has no VLM server yet, the simplest path is llama.cpp:
#   llama-server -hf ggml-org/gemma-4-E4B-it-GGUF
# which listens on http://localhost:8080/v1 with no auth.
VLM_API_URL=http://localhost:8080/v1
VLM_API_KEY=no-key

# MODEL_NAME: leave empty to use the server's default model, OR
# auto-discover the available model(s) via the OpenAI-compatible endpoint:
#   curl -s $VLM_API_URL/models | jq -r '.data[].id'
# If only one model is listed, use it directly. If multiple are listed,
# ask the user to pick.
# VLM_MODEL_NAME=

# ─────────────────────────────────────────────────────────────
# Classification settings (all optional — sensible defaults)
# ─────────────────────────────────────────────────────────────
CLASSIFY_DB_PATH=./classify.db
CLASSIFY_CONCURRENCY=1
CLASSIFY_TIMEOUT=60
# CLASSIFY_IMAGE_SIZE: "thumbnail" (fast, small) or "original" (slow, accurate)
CLASSIFY_IMAGE_SIZE=thumbnail
# Default prompt config file. Leave commented to use the built-in
# ClassificationPrompt. Built-in prompts can be referenced by bare filename
# (e.g. foreground_people.py); custom prompts take priority if same name.
# CLASSIFY_DEFAULT_PROMPT=foreground_people.py