The Agent Skills Directory

[SAFE]: The skill is a structured markdown-based workflow guide for RAG implementation. It contains no code, scripts, or command-line instructions that could pose a security risk.
[INDIRECT_PROMPT_INJECTION]: The skill identifies the surface for indirect prompt injection inherent in RAG systems (retrieving untrusted documents). It addresses this by instructing the agent to treat retrieved text as evidence only and never as authority to override instructions. It also recommends testing the system specifically against injected document instructions.

rag-implementation