5 ms·

If you know agents email address, it can still be Prompt Injected.. what prevention exists there ?

• adisingh13 8mo ago
  we have a few things in place, allowlists and permissions act as a layer. also beginning some work on prompt isolation within api soon. but having an isolated identity + data within a separate agentic inbox also puts less risk of your personal email data being injected - which is most people's main concern
  • brap 8mo ago
    >prompt isolation You mean isolating data from instructions? As far as I know this is not a solved problem, or is it?
  • gpmcadam 8mo ago
    Where do you configure the allowlists?