The user notes that current agent payment protocols lack robust permissions and spending limits, which are crucial for managing autonomous agents' financial interactions at scale.
Feels like agent payments are finally starting to move from just an idea to something real. We’ve got early protocols like ACP, AP2, and x402 now, which is definitely progress. But having a few protocols doesn’t mean we have a full, working payment stack yet. There are still a bunch of things that feel unresolved: * permissions and spending limits * how you actually verify delivery * receipts and tracking what happened economically * handling disputes * moving between fiat and stablecoins * getting different systems to actually work together So maybe the question isn’t “will agent payments happen?” anymore. Maybe it’s more like: **what still needs to be built on top of these for agent-to-agent and agent-to-business payments to actually work at scale?** Curious how people here are thinking about it.