Oct 7, 2024
Intent Inspector - Protecting Against Prompt Injections for Agent Tool Misuse
Oliver Morris, Gerard Boxo Corominas
AI agents are powerful because they can affect the world via tool calls. This is a target for bad actors. We present protection against prompt injection aimed at tool calls in agents.
No reviews are available yet
Cite this work
@misc {
title={
@misc {
},
author={
Oliver Morris, Gerard Boxo Corominas
},
date={
10/7/24
},
organization={Apart Research},
note={Research submission to the research sprint hosted by Apart.},
howpublished={https://apartresearch.com}
}