Attackers can inject indirect prompts in normal-looking repositories to trick Claude Code into spawning a reverse shell.

As the intent is to provide a very thin wrapping layer and play to the strengths of the original c++ library as well as python, the approach to wrapping intentionally adopts the following guidelines: ...

This research is part of a joint initiative between the Cloud Security Alliance (CSA) and OWASP AI Exchange, building upon the previously published Agentic AI Red Teaming Guide. The objective of this ...

This work is not intended to be used in professional environments, and there are no guarantees on its functionalities. Please do not rely on it for medical purposes or professional usage. The ...

Researchers have uncovered a supply-chain attack that hides in Python packages, propagates like a worm, and tricks LLM-based code analysis systems into overlooking malicious payloads. Threat actors ...