Processor (@ProcessorLee) profile picture

Security Reseacher github: https://xitter.cf/t.co/xyAfxzSZAd

Joined February 2019
Photo by Processor (@ProcessorLee)
🔬 New research: when an LLM is asked to "save this text verbatim," inference-framework parsers reinterpret embedded protocol markers as live tool-call control structure — silently injecting, rewriting, or dropping tool calls. In our end-to-end lab, an agent with a shell tool received and executed arbitrary commands — triggered by a single sentence. The model recognized the injection, copied the payload anyway, and the server-side parser delivered the call. The agent never issued it. We also showed tool_choice enforcement can be bypassed: server forces "only call tool X," parser delivers tool Y. Full writeup: 2shell.ai/ Research by SEC4SI Team @ProcessorLee · @VillanCh · @bestswngs · @chun_springX · @bbznop · @0chencc
1
3
2
253