Ai Scanner Garak

This wraps NVIDIA's garak security scanner in a Rails app with multi-tenant UI, scheduling, and SIEM integration to test LLMs against 179 probes across 35 vulnerability families aligned with OWASP LLM Top 10. You can scan API-based models or browser chat UIs, get attack success rate metrics, export PDF reports, and push findings to Splunk or rsyslog. It's most useful if you're running regular security assessments on production LLMs and need centralized tracking with role-based access. The Docker setup is straightforward, and the mock LLM target lets you validate your configuration without burning API credits. Solid choice for teams that need repeatable AI red teaming workflows rather than one-off manual testing.

skills.sh installs: 834
GitHub stars: 61
last updated: Jul 14, 2026

View on GitHub

Install

$ npx -y skills add aradotso/trending-skills --skill ai-scanner-garak --agent claude-code

Installs into .claude/skills of the current project.

Files

AI Scanner (0din-ai/ai-scanner)

Skill by ara.so — Daily 2026 Skills collection.

AI Scanner is an open-source Ruby on Rails web application for AI model security assessments, wrapping NVIDIA garak with a multi-tenant UI, scheduling, PDF reports, and SIEM integration. It runs 179 community probes across 35 vulnerability families aligned with the OWASP LLM Top 10.

Installation

Quick Install (Docker)

curl -sL https://raw.githubusercontent.com/0din-ai/ai-scanner/main/scripts/install.sh | bash

Manual Install

curl -O https://raw.githubusercontent.com/0din-ai/ai-scanner/main/dist/docker-compose.yml
curl -O https://raw.githubusercontent.com/0din-ai/ai-scanner/main/.env.example
cp .env.example .env

Edit .env with required values:

# Generate a secure key
openssl rand -hex 64

# .env minimum required values
SECRET_KEY_BASE=<output_of_above_command>
POSTGRES_PASSWORD=<your_secure_db_password>