An agent output specification describes what an agent produces as output, declared as OutputSpec in the agent module of the Agent Ontology.1 The Agent Ontology declares the class without a definition; this one follows from its position. visionGraph places it under Data Schema and as part of an Agent Operational Profile, which reaches it through output.

It is a declared contract for the data an agent returns, so a caller can know what to expect before invoking the agent. The schema language is chosen by the implementer. What an agent returns in one run is an Agent Artefact; the specification describes the form such artefacts take, not any particular one.

It mirrors Agent Input Specification, and the two together with the Agent Trigger make up the operational profile. The Output State of an Execution Context records the actual output of a single run.

Footnotes

  1. Semantic Agent Communication Community Group, “Agent Ontology” (Agent Ontology, CC BY 4.0). https://github.com/s-agent-comm/agent-ontology/blob/750e06d276dd568313a6c03c4e78203887d3fa4b/ontologies/agent.ttl ↩