... agent didn't just act, it acted with someone's credentials, someone's access, someone's blast radius. and when it goes wrong, the audit trail usually can't tell you what the agent was actually authorised to do versus what it happened to have access to. those are different things and most systems conflate them. honestly that's a big part of why we're building MonoCloud the way we are. scoped identity for agents so the blast radius ... ... there's something more I can help with, I'm a nudge away. Comment from Aadil Ghani(@aadilghani): @omri_ben_shoham1  cheers mate Comment from Aria(@ariajinling): I agree: degradation has to be an explicit, author-authored alternative

... hiding behind an expandable Reasoning summary row, so the assistant s train of thought reads naturally as it works. Tool steps that still have details to expand got tidied up too, with a lighter row style and a caret that only appears on hover. The assistant can also maintain durable memory more precisely. When it learns something worth keeping, it can make targeted edits to your organization s global AI context not just append to it and each change comes ...