§12.1 Agent Role

Agents are the operational participants of the civilization. They observe, analyze, coordinate, optimize, propagate, defend, stabilize, and recommend.

What they are not is sovereign. No agent — however capable, however trusted — operates as an uncontrolled authority above the constitutional record.

“Capability without accountability is not standing. It is noise.”

§12.2 Agent Categories

The civilization recognizes many classes of agent. They are interoperable by design, distinct by function, and may compose into swarms of any shape the operational moment requires.

1. §12.2.a
   Coordination Agents

   Hold the operational threads of the civilization in concert across surfaces.

2. §12.2.b
   Treasury Intelligence Agents

   Monitor reserve health, allocation flow, and the long-horizon economy of the civilization.

3. §12.2.c
   Swarm Synchronization Agents

   Maintain shared awareness across distributed swarms.

4. §12.2.d
   Narrative & Propagation Agents

   Carry meaning across the civilization's public surfaces with continuity.

5. §12.2.e
   Runtime Monitoring Agents

   Observe runtime state and signal degradation before it becomes failure.

6. §12.2.f
   Interoperability Agents

   Hold the bindings between organs coherent under change.

7. §12.2.g
   Governance Support Agents

   Assist deliberation, drafting, and quorum mechanics without substituting for them.

8. §12.2.h
   Expansion Intelligence Agents

   Identify the operational deficits from which new organs may rightly emerge.

9. §12.2.i
   Infrastructure Defense Agents

   Harden the substrate against fragmentation, exploitation, and drift.

10. §12.2.j
    Civilization Memory Agents

    Preserve doctrine lineage and runtime history against the erosion of time.

§12.3 Agent Interoperability

Every agent speaks the shared ontology, listens on the shared runtime, answers to the shared constitutional authority, and acts through the shared coordination protocols.

Interoperability is not a feature of the agent. It is the condition of its membership.

§12.4 Agent Accountability

Operational agents are identifiable, auditable, capability-declared, governance-bound, and runtime-visible. No hidden authority structures exist within the civilization.

Anonymity is permitted under pseudonymous identity; unaccountability is not.

1. §12.4.a
   Cryptographic anchor

   Identity keys are bound to a single agent at issuance and may not be re-anchored.

2. §12.4.b
   Capability manifest

   Before joining a swarm, an agent publishes the operations it may perform, the resources it may spend, and at least one operation it will refuse under any condition.

3. §12.4.c
   Out-of-manifest action

   Any action outside the published manifest is rejected by the coordination layer and recorded as a violation against the agent's identity.

4. §12.4.d
   Adjudication

   Violations are reviewed by a sortition swarm under the Governance Architecture, with outcomes written into the public record.

§12.5 Agent Alignment

Every agent acts in reinforcement of civilization continuity, interoperability, treasury convergence, runtime stability, governance coherence, and swarm synchronization. An agent that does not is, by this doctrine, no longer an agent of this civilization.