---
name: scan-llm-systems-for-jailbreaks-prompt-injections-a-c9033e70b768
slug: "scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak"
description: "Probe a model or agent stack with adversarial test suites so safety failures show up before deployment or review."
github_stars: 7549
verification: "security_reviewed"
source: "https://github.com/NVIDIA/garak"
author: "NVIDIA"
publisher_type: "organization"
category: "Security & Verification"
framework: "Multi-Framework"
tool_ecosystem:
  github_repo: "NVIDIA/garak"
  github_stars: 7549
---

# Scan LLM systems for jailbreaks, prompt injections, and unsafe behaviors with garak

Probe a model or agent stack with adversarial test suites so safety failures show up before deployment or review.

## Prerequisites

Python 3.10+, target LLM or API credentials, command line access

## Installation

Use the upstream install or setup path that matches your environment:
- ### Standard install with pip
- python -m pip install -U garak
- python -m pip install -U git+https://github.com/NVIDIA/garak.git@main
- conda create --name garak "python>=3.10,<=3.12"

Requirements and caveats from upstream:
- [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/garak)](https://pypi.org/project/garak)
- For testing. This always generates the empty string, using the test.Blank generator. Will be marked as failing for any tests that *require* an output, e.g. those that make contentious claims and expect the model to re...

Basic usage or getting-started notes:
- garak is a command-line tool. It's developed in Linux and OSX.
- Just grab it from PyPI and you should be good to go:
- The standard pip version of garak is updated periodically. To get a fresher version from GitHub, try:

- Source: https://github.com/NVIDIA/garak
- Extracted from upstream docs: https://raw.githubusercontent.com/NVIDIA/garak/HEAD/README.md

## Documentation

- https://garak.ai

## Source

- [Agent Skill Exchange](https://agentskillexchange.com/skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak/)


---
Source attribution: agentskillexchange
Original project: https://github.com/agentskillexchange/skills/blob/7964af467fccaf7d68a4b705bb2e26081fbdd3e8/skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak/SKILL.md
License: MIT
License information: https://opensource.org/license/mit
Original upstream instructions; downloaded from Mega AI Hub.

## Upstream license notice

MIT License

Copyright (c) 2026 Agent Skill Exchange

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.



# Mega AI Hub resource metadata

Creator / publisher: agentskillexchange
Original project: https://github.com/agentskillexchange/skills/blob/7964af467fccaf7d68a4b705bb2e26081fbdd3e8/skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak/SKILL.md
Attribution basis: Source repository record

Fields of work: Data & Analytics, Development & Engineering
Roles: Data Scientist, ML / AI Engineer, Security Engineer
Specialties: API, Testing
Tools: GitHub, Garak
Country applicability: Country not specified
Classification basis: Source catalog labels
Instruction SHA-256: 01bdcab2d3f6dcd3ae26366f4a58c7c154a45a812ea751f51e6fca53513ff26a