bettaher_adam

3 ms·

1. 1.▲
   by bettaher_adam 6mo ago

   The fail-closed approach is the right default. One thing I'd add to the attack classes you're considering: prompt injection via filesystem reads — an attacker can craft a file that, when read by the agent, injects instructions
2. 2.▲
   by bettaher_adam 6mo ago

   The skeptic loop concept is the most interesting part here. One thing worth considering for the validation layer: signing the intermediate outputs between agents with HMAC so you can trace exactly which agent in the chain produced which
3. 3.▲
   by bettaher_adam 6mo ago

   Interesting approach to knowledge unit validation. One thing I've noticed when building constrained LLM pipelines: separating the system prompt from user input at the message level (not string concatenation) makes a significant dif