News for hackers. Stuff that matters.
Re:Design Patterns for Securing LLM Agents Against Prompt In... by babyshake on Saturday June 14, 2025 @03:30PM (#44278291)
One of the tricky things is untrusted input somehow making its way into what is otherwise considered trusted input. There are obviously untrusted inputs like a customer support chatbot. And there are maybe trusted inputs, like a codebase that probably doesn't contain harmful instructions, but there's always a chance that harmful instructions might be able to make their way into it.

There are no comments on this comment yet.