# ai-prompt-injection

> Test LLM-backed apps for prompt injection (direct + indirect) and its consequences: data exfil, tool/function abuse, guardrail bypass. Load when the target is a chatbot/assistant/ agent, summarizes untrusted content, has tools/functions, or does RAG. Signals: "ask AI", system prompts, function-calli

Skill · Grátis · Licença: MIT · Versão: 2026.10.14-ee0e0c5

Instalar com a CLI da NotHoly: `notholy skills install noorqureshi/ai-prompt-injection`

Canonical: https://notholy.com/pt-br/skills/noorqureshi/ai-prompt-injection
