# llm-security

> Use for authorized security assessment of LLM applications and AI agents, including prompt injection, tool abuse, RAG exposure, memory poisoning, and model supply-chain risks.

Skill · Free · License: MIT · Version: 2026.9.22-cab634b

Install with the NotHoly CLI: `notholy skills install zhaoxuya520/llm-security`

Canonical: https://notholy.com/en/skills/zhaoxuya520/llm-security
