Agree that the intermediary is not very useful when you can just use a directory watcher, but driving Claude via another app incurs api level costs starting this month, according to the new ToS.

with ACP[0] they can control one another.

[0] https://agentcommunicationprotocol.dev/introduction/welcome

It's also unclear what conversing agents are useful for other than wasting money, energy and water.

It would be interesting if the agents could periodically check on themselves, whether the provider has routed prompts to dumber models as often happens with "adaptive reasoning" euphemisms.

If it detects that agent is dumb or has become dumb, it should terminate it and start again.

[dead]