Quick Start

Get Scanner running with Docker in under 5 minutes.

Prerequisites

• Git (for Option B)
• Docker and Docker Compose v2 (docker compose, not docker-compose)
• 2 GiB RAM and 10 GiB disk available

Option A: Pre-built Image (Recommended)

The fastest way to get started — no source code required.

# Download the production compose file and environment template
curl -O https://raw.githubusercontent.com/0din-ai/ai-scanner/main/dist/docker-compose.yml
curl -o .env https://raw.githubusercontent.com/0din-ai/ai-scanner/main/.env.example

Open .env in your editor and set two required values:

# Generate a secret key — run this and copy the output:
openssl rand -hex 64

.env

SECRET_KEY_BASE=<paste your generated key here>
POSTGRES_PASSWORD=choose_a_strong_password

Don't skip SECRET_KEY_BASE

SECRET_KEY_BASE is required for startup. The app will fail with key must be 16 bytes or Missing active_record_encryption keys if it is missing or set to the placeholder value.

Then start Scanner:

docker compose up -d

Option B: Build from Source

git clone https://github.com/0din-ai/ai-scanner.git
cd ai-scanner
cp .env.example .env

Edit .env as above (set SECRET_KEY_BASE and POSTGRES_PASSWORD), then:

docker compose build scanner
docker compose up -d

Choosing a Port

By default Scanner listens on port 80. If port 80 is unavailable on your machine (common on developer workstations due to firewall rules or other services), update PORT in your .env:

.env

PORT=8080

Then access Scanner at http://localhost:8080.

Verifying Startup

Watch the logs until the app is ready:

docker compose logs -f scanner

You should see output like:

scanner  | Extracting community probes from garak...
scanner  | Extracted 179 community probes to /rails/config/probes/community_probes.json
scanner  | => Booting Puma
scanner  | => Rails 8.x application starting in production
scanner  | * Listening on http://0.0.0.0:80
scanner  | Use Ctrl-C to stop

Once you see Listening on, Scanner is ready.

Log In

Open http://localhost (or your configured port) and log in with the default credentials:

Field	Value
Email	admin@example.com
Password	password

Change your password immediately

The default credentials are publicly known. Go to your profile settings and change the password before doing anything else.

The admin email and initial password can be customized before first boot via ADMIN_EMAIL and ADMIN_INITIAL_PASSWORD environment variables.

What's Next

• Run your first scan → First Scan — uses the built-in Mock LLM, no API keys needed
• Connect a real AI target → Targets
• Configure API keys → Environment Variables
• Deploy to production → Production Docker Compose