Advanced · Systems

Agent Capability Vocabulary

Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.

TL;DR. A shared, standardized naming scheme for the actions agents can offer and request.

Technical Definition

A shared, standardized naming scheme for the actions agents can offer and request.

How it works

A capability vocabulary defines canonical names and schemas for actions — for example a payment capability with its parameters, limits, and effects — so that an agent advertising a capability and an agent requesting it mean the same thing. Vocabularies are usually namespaced, versioned, and hierarchical so that authorization can be granted at a coarse level and attenuated to a finer one. They are the interoperability substrate for discovery, authorization, and intent matching.

Related Concepts

• Intent Description Language (IDL) — A structured, machine-readable format for expressing what an agent intends to do, for whom, and under what constraints.
• Agent Capabilities — The concrete set of actions an agent can perform — the tools, APIs, and side effects available to it.
• Capability-Based Agent Authorization — Authorizing agents by handing them unforgeable, narrowly scoped capability tokens rather than checking roles at call time.
• Agent Protocol Semantics — The shared meaning layer of an agent protocol — what messages, verbs, and capability names actually denote.