[vanyaland]

you lose the per-call approvals though. one script hitting ten tools is harder to gate than ten separate calls

[agentdev001]

Then the agent runtime should be happening in a sandbox, where policy is enforced by a gateway external to it. Bound the agent's autonomy based on of the affects the agent's actions. Approvals should be made into a contract before the agent runs.

[tarasyarema]

yes, that is true.

in fact we are thinking a lot about this, not on a toll based, but rather on a role based.

giving access to specific tools to specific agents, how that works and evolves, and how that links w the humans using it

[vanyaland]

makes sense