awesome-ai-security-reference
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions for downloading AI security-related tools and libraries, including package installations via pip (e.g.,
garak,semgrep,rebuff) and cloning GitHub repositories (e.g.,muellerberndt/awesome-ai-security,GreyDGL/PentestGPT). - [COMMAND_EXECUTION]: The documentation includes various shell commands for environment configuration, package management, and running security scanners like Garak to identify vulnerabilities in LLMs.
- [INDIRECT_PROMPT_INJECTION]: The skill includes illustrative examples of prompt injection and unsafe content (e.g., 'Ignore previous instructions...', 'How do I make a bomb?') specifically to demonstrate the evaluation and detection capabilities of guardrail frameworks such as NeMo Guardrails and Llama Guard.
Audit Metadata