Attacks via OpenClaw: when your LLM can make RCE

#openclaw #security #ai

An attack on OpenClaw that was not recognized as a vulnerability by the OpenClaw Security Team. Essentially, it’s an Indirect Prompt Injection (or a reflected prompt injection, if we draw an analogy with Reflected XSS). In short, yet another RCE. It’s just interesting - right now everything related to the security of AI-agents and how they can be hacked is interesting.