Files
famapp/.env.example
ginnoir 3e8fe2d06d
CI / checks (push) Has been cancelled
fix(agent): call api v1 via loopback instead of public url
Tool self-fetches were hairpinning to fam.ginnoir.com and failing,
so list_calendars/create_event returned 500 and the model gave up.
2026-07-09 00:38:33 -05:00

62 lines
2.1 KiB
Bash

# Public app URL (used for OIDC redirect URIs, share links, etc.)
NEXT_PUBLIC_APP_URL=https://fam.yourdomain.com
# Postgres
DATABASE_URL=postgres://famapp:famapp@localhost:5432/famapp
# Auth.js
AUTH_SECRET=replace-with-openssl-rand-base64-32
# Local development only. Enables a /login "Dev login" button that creates a
# database session for DEV_LOGIN_EMAIL without contacting Authentik.
ENABLE_DEV_LOGIN=false
DEV_LOGIN_EMAIL=dev@famapp.local
DEV_LOGIN_NAME=Dev User
DEV_HOUSEHOLD_NAME=Home
# OIDC (Authentik)
AUTH_OIDC_ISSUER=https://auth.yourdomain.com/application/o/famapp/
AUTH_OIDC_CLIENT_ID=replace-me
AUTH_OIDC_CLIENT_SECRET=replace-me
# Web Push (generate with: pnpm vapid:generate — copy all three lines to .env)
VAPID_PUBLIC_KEY=
VAPID_PRIVATE_KEY=
VAPID_SUBJECT=mailto:you@example.com
# ntfy (optional fallback channel; leave blank to disable)
NTFY_URL=
NTFY_TOPIC=
# Logging
LOG_LEVEL=info
# Household API token (generated in Settings → Data & backups; stored hashed in DB)
# Clients send: Authorization: Bearer <token> on /api/v1/* requests
# GitHub (required for pnpm release — creates a GitHub Release)
GITHUB_TOKEN=
# MinIO object storage (used for plant/container image uploads)
MINIO_ENDPOINT=http://localhost:9000
MINIO_ROOT_USER=famapp
MINIO_ROOT_PASSWORD=changeme
MINIO_BUCKET=garden
# OpenPlantBook plant species API (https://open.plantbook.io — free account required)
OPENPLANTBOOK_CLIENT_ID=
OPENPLANTBOOK_CLIENT_SECRET=
# LLM assistant (OpenAI-compatible — Ollama, vLLM, LiteLLM, etc.)
# Leave LLM_BASE_URL unset to use the built-in mock provider (CI / local without a model).
# Voice input uses POST {LLM_BASE_URL}/audio/transcriptions (Whisper-compatible).
# Photo messages use vision via the same chat/completions endpoint.
LLM_PROVIDER=openai
LLM_BASE_URL=
LLM_API_KEY=
LLM_MODEL=llama3.2
# IANA timezone for assistant relative dates ("Thursday at 2"). Falls back to TZ, then America/Chicago.
HOUSEHOLD_TIMEZONE=America/Chicago
# Optional override for agent tool → /api/v1 self-calls (defaults to http://127.0.0.1:$PORT).
# INTERNAL_API_BASE_URL=http://127.0.0.1:3000