More Agents Inside, Fewer Agents on the Network

If agents work with other agents, they need a standard way to talk. That was the case for an explicit agent-to-agent protocol. Adoption is real. Meanwhile the rest of the stack has reduced how often agents need to appear to one another as independently addressable peers.

A remote agent can look like a capability. Inference stays behind the interface. Subagents live inside one runtime and die with the task. A supervisor presents one identity while specialists stay internal. Software factories coordinate the work, not a mesh of persistent agent relationships. Workflow engines treat agents as steps in a graph they already own. Handoffs can be artifacts in durable state rather than messages. Queues and event buses already do this at scale.

The strongest case for an agent-to-agent protocol is across organizational boundaries. Even there, the remote system may not need to expose an agent. It may need to expose a stable capability. The pressure is architectural: the winning pattern is often to hide agents, not to put them on the network.

This Ledger Entry expands how the reader thinks by replacing “agents must talk to agents” with “coordination can live in capabilities, runtimes, work objects, and existing distributed systems.”