The primary semantic coordinate for defining and governing how AI agents behave.
A clean two-token .ai position at the intersection of agent design and behavioral specification.
Coordinated sets this position belongs to — the coverage it extends. Counts are the live cluster size in the graph.
Architectural context
Agent · Cross-Vertical · 2 compound moats. Architectural surface: Agent. Cross-cutting: Behavior.
Layer position: Cross-cutting
Why this is canonical
'Agent behavior' is the operative phrase across safety research, reinforcement learning literature, and agentic systems design — it names what practitioners actually specify, test, and constrain when building autonomous systems. The .ai TLD aligns it directly with the agent era.
Where it fits
A few directions this coordinate opens —
Illustrative, not exhaustive — held as a transferable canonical position, open to the buyer's own use.