Module agent 

Agent-identifier model (#16): resolve an identifier’s AGENT TYPE and check grounding.

An agent’s type is carried as a graph relation (rdf:type → a values agent class), the same modal-predicate pattern as crate::modal_kind, resolved via zero-alloc QuinIndex point lookups (one identity space, #14). On top of resolution this activates the agency.n3 grounding guard at runtime: an ArtificialAgent / PlatformAgent acting with no Principal (values:operatedBy) is UngroundedAgency — the G1’ guard that keeps an AI agent accountable to a human principal rather than free-floating. (Agent identity is still relational + never definitive — this resolves a declared type, it does not fix the agent; see principle-identifiers-not-identity.)

Constants

A_ARTIFICIAL_AGENT

A_LEGAL_PERSON

A_NATURAL_PERSON

A_PLATFORM_AGENT

A_PUBLIC_AUTHORITY

P_OPERATED_BY

The accountable principal behind an agent (agency.n3 values:operatedBy).

P_RDF_TYPE

Functions

agent_type

The declared agent class of agent (its rdf:type), if any.

agent_type_name

Readable name for a known agent class.

is_artificial

Whether an agent class is a non-personhood artificial agent (must be grounded).

is_ungrounded_agency

agency.n3 G1’ grounding guard: an ArtificialAgent / PlatformAgent acting with no Principal is UngroundedAgency. Returns true when the agent trips the flag.

principal_of

The accountable principal behind agent (values:operatedBy), if declared.