0din-ai/ai-scanner

AI model safety scanner built on NVIDIA garak

What it solves

It provides a way for organizations to perform security assessments on AI models before they are deployed. It acts as a penetration testing tool for AI systems, identifying vulnerabilities like jailbreaks and other security flaws that could be exploited in production.

How it works

Built with Ruby on Rails and powered by NVIDIA garak, the application runs a series of "probes" (attack prompts) against AI targets. These targets can be API-based LLMs or browser-based chat interfaces. The system tracks the Attack Success Rate (ASR) and provides detailed logs and reports on which attacks succeeded.

Who it’s for

Security teams and organizations deploying AI models who need to validate the safety and robustness of their systems against known threat patterns.

Highlights

  • Extensive Probe Library: Includes 179 community probes across 35 vulnerability families, aligned with the OWASP LLM Top 10.
  • Real-World Attacks: Ships with 0DIN-disclosed jailbreaks and multiple retargetable variants rather than simulations.
  • Enterprise Features: Supports multi-tenant deployments, scheduled scanning, and SIEM integration (Splunk, Rsyslog).
  • Detailed Reporting: Offers PDF exports with per-probe drill-downs and trend tracking for ASR scoring.

Related

  • Project
    msoedov/agentic_security

    An open-source vulnerability scanner for LLMs and agent workflows that protects AI systems from jailbreaks, fuzzing, and multimodal attacks.

  • Project
    Tencent/AI-Infra-Guard

    AI‑Infra‑Guard (A.I.G) is an open‑source, Docker‑deployable platform for red‑team security testing of AI services. It scans live LLM endpoints, AI‑agent workflows, and model‑control‑plane code for CVEs, jailbreak susceptibility, and agent loss‑of‑control risks, exposing results via a web UI and REST API.

  • Project
    NVIDIA/garak

    garak is an open‑source CLI framework that automatically probes large language models for security‑relevant failures (hallucination, prompt‑injection, data leakage, toxicity, jailbreaks, etc.). It supports Hugging Face, OpenAI, Bedrock, NIM, Replicate, local GGUF, and generic REST models, and provides a large library of ready‑made probes and detectors. Results are shown as PASS/FAIL tables and saved as JSONL logs for deeper analysis.

  • Project
    confident-ai/deepteam

    An open-source red teaming framework for LLM systems that simulates adversarial attacks to uncover vulnerabilities and provides guardrails for production security.

  • Project
    cisco-ai-defense/skill-scanner

    Skill Scanner is an open‑source security scanner for AI‑agent skill packages (e.g., OpenAI Codex, Cursor, Claude). It combines signature‑based YARA/YAML rules, AST‑based data‑flow analysis, optional LLM‑as‑a‑judge, and a CEL decision layer to detect prompt‑injection, data‑exfiltration, and malicious code patterns. Install via pip, run from the CLI or Python SDK, and integrate with CI/CD via GitHub Actions or a pre‑commit hook. Findings are reported with severity levels; a clean scan is best‑effort only and does not guarantee safety.