diff --git a/.env.example b/.env.example
index 350cb34..e743c44 100644
--- a/.env.example
+++ b/.env.example
@@ -1,2 +1,21 @@
-# Copy this to .env and fill in your values
-# Not currently used — configuration is in Caddyfile and docker-compose.yml
+# Copy this file to .env and fill in your values.
+# docker compose reads .env automatically — no need to edit docker-compose.yml.
+
+# ── Security ──────────────────────────────────────────────────────────────
+# Long random string used to sign login tokens. Change this before going live.
+# Generate one with: openssl rand -hex 32
+JWT_SECRET=change-me-in-production
+
+# Password for the /admin panel.
+ADMIN_PASSWORD=admin
+
+# ── AI Writing Prompts (optional) ─────────────────────────────────────────
+# Base URL of your Ollama / OpenAI-compatible server (no trailing slash).
+# The app will call POST ${OLLAMA_URL}/v1/chat/completions
+OLLAMA_URL=https://ai.binarygnome.com
+
+# Model name as it appears on your server.
+OLLAMA_MODEL=gemma4:latest
+
+# API key / Bearer token — leave blank if your server doesn't require one.
+OLLAMA_API_KEY=