Agent Security Is a Systems Problem
A new paper argues that LLM agent security cannot be solved by better prompts or guardrails alone. The architecture itself leaks privilege.
Read the note 1 min read
From the notebook
sandboxing.
1 note on this topic.