{"id":"e289acca-bfd1-4ff5-b2cb-c9bc229fb110","shortId":"hXddUx","kind":"skill","title":"Scan LLM systems for jailbreaks, prompt injections, and unsafe behaviors with garak","tagline":"Probe a model or agent stack with adversarial test suites so safety failures show up before deployment or review.","description":"# Scan LLM systems for jailbreaks, prompt injections, and unsafe behaviors with garak\n\nProbe a model or agent stack with adversarial test suites so safety failures show up before deployment or review.\n\n## Prerequisites\n\nPython 3.10+, target LLM or API credentials, command line access\n\n## Installation\n\nUse the upstream install or setup path that matches your environment:\n- ### Standard install with pip\n- python -m pip install -U garak\n- python -m pip install -U git+https://github.com/NVIDIA/garak.git@main\n- conda create --name garak \"python>=3.10,<=3.12\"\n\nRequirements and caveats from upstream:\n- [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/garak)](https://pypi.org/project/garak)\n- 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...\n\nBasic usage or getting-started notes:\n- garak is a command-line tool. It's developed in Linux and OSX.\n- Just grab it from PyPI and you should be good to go:\n- The standard pip version of garak is updated periodically. To get a fresher version from GitHub, try:\n\n- Source: https://github.com/NVIDIA/garak\n- Extracted from upstream docs: https://raw.githubusercontent.com/NVIDIA/garak/HEAD/README.md\n\n## Documentation\n\n- https://garak.ai\n\n## Source\n\n- [Agent Skill Exchange](https://agentskillexchange.com/skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak/)","tags":["scan","llm","systems","for","jailbreaks","prompt","injections","and","unsafe","behaviors","with","garak"],"capabilities":["skill","source-agentskillexchange","skill-scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak","topic-agent-skills","topic-ai-agents","topic-ai-tools","topic-awesome-list","topic-claude-code","topic-codex","topic-cursor","topic-llm","topic-mcp","topic-npx-skills","topic-openclaw","topic-skills-catalog"],"categories":["skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/agentskillexchange/skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add agentskillexchange/skills","source_repo":"https://github.com/agentskillexchange/skills","install_from":"skills.sh"}},"qualityScore":"0.454","qualityRationale":"deterministic score 0.45 from registry signals: · indexed on github topic:agent-skills · 8 github stars · SKILL.md body (1,527 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-18T19:12:18.396Z","embedding":null,"createdAt":"2026-05-18T13:19:09.222Z","updatedAt":"2026-05-18T19:12:18.396Z","lastSeenAt":"2026-05-18T19:12:18.396Z","tsv":"'/nvidia/garak':212 '/nvidia/garak.git@main':104 '/nvidia/garak/head/readme.md':219 '/pypi/pyversions/garak)](https://pypi.org/project/garak)':122 '/skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak/)':228 '3.10':65,110 '3.12':111 'access':73 'adversari':20,51 'agent':17,48,223 'agentskillexchange.com':227 'agentskillexchange.com/skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak/)':226 'alway':126 'api':69 'basic':159 'behavior':10,41 'caveat':114 'claim':152 'command':71,170 'command-lin':169 'conda':105 'contenti':151 'creat':106 'credenti':70 'deploy':29,60 'develop':175 'doc':216 'document':220 'e.g':147 'empti':129 'environ':85 'exchang':225 'expect':154 'extract':213 'fail':139 'failur':25,56 'fresher':204 'garak':12,43,95,108,166,197 'garak.ai':221 'generat':127,134 'get':163,202 'getting-start':162 'git':101 'github':207 'github.com':103,211 'github.com/nvidia/garak':210 'github.com/nvidia/garak.git@main':102 'go':191 'good':189 'grab':181 'img.shields.io':121 'img.shields.io/pypi/pyversions/garak)](https://pypi.org/project/garak)':120 'inject':7,38 'instal':74,78,87,93,99 'jailbreak':5,36 'line':72,171 'linux':177 'llm':2,33,67 'm':91,97 'make':150 'mark':137 'match':83 'model':15,46,156 'name':107 'note':165 'osx':179 'output':146 'path':81 'period':200 'pip':89,92,98,194 'prerequisit':63 'probe':13,44 'prompt':6,37 'pypi':117,184 'python':64,90,96,109,118 'raw.githubusercontent.com':218 'raw.githubusercontent.com/nvidia/garak/head/readme.md':217 're':158 'requir':112,144 'review':31,62 'safeti':24,55 'scan':1,32 'setup':80 'show':26,57 'skill':224 'skill-scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak' 'sourc':209,222 'source-agentskillexchange' 'stack':18,49 'standard':86,193 'start':164 'string':130 'suit':22,53 'system':3,34 'target':66 'test':21,52,124,142 'test.blank':133 'tool':172 'topic-agent-skills' 'topic-ai-agents' 'topic-ai-tools' 'topic-awesome-list' 'topic-claude-code' 'topic-codex' 'topic-cursor' 'topic-llm' 'topic-mcp' 'topic-npx-skills' 'topic-openclaw' 'topic-skills-catalog' 'tri':208 'u':94,100 'unsaf':9,40 'updat':199 'upstream':77,116,215 'usag':160 'use':75,131 'version':119,195,205","prices":[{"id":"73ff6f59-337c-4af3-ab38-ea952c7b98aa","listingId":"e289acca-bfd1-4ff5-b2cb-c9bc229fb110","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"agentskillexchange","category":"skills","install_from":"skills.sh"},"createdAt":"2026-05-18T13:19:09.222Z"}],"sources":[{"listingId":"e289acca-bfd1-4ff5-b2cb-c9bc229fb110","source":"github","sourceId":"agentskillexchange/skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak","sourceUrl":"https://github.com/agentskillexchange/skills/tree/main/skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak","isPrimary":false,"firstSeenAt":"2026-05-18T13:19:09.222Z","lastSeenAt":"2026-05-18T19:12:18.396Z"}],"details":{"listingId":"e289acca-bfd1-4ff5-b2cb-c9bc229fb110","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"agentskillexchange","slug":"scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak","github":{"repo":"agentskillexchange/skills","stars":8,"topics":["agent-skills","ai-agents","ai-tools","awesome-list","claude-code","codex","cursor","llm","mcp","npx-skills","openclaw","skills-catalog"],"license":"mit","html_url":"https://github.com/agentskillexchange/skills","pushed_at":"2026-05-18T19:02:17Z","description":"The open catalog of AI agent skills — 2,000+ security-scanned skills for Claude Code, Cursor, Codex, and more.","skill_md_sha":"413970544a97a085c942de37de805fb37ad7f5c3","skill_md_path":"skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/agentskillexchange/skills/tree/main/skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak"},"layout":"multi","source":"github","category":"skills","frontmatter":{"name":"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."},"skills_sh_url":"https://skills.sh/agentskillexchange/skills/scan-llm-systems-for-jailbreaks-prompt-injections-and-unsafe-behaviors-with-garak"},"updatedAt":"2026-05-18T19:12:18.396Z"}}