Reception Infrastructure for Conversational Agents: _Consent, Memory Provenance, and Runtime Governance in Local LLM Systems_

We present a reproducible runtime governance architecture for local conversational agents. The system combines consent scaffolding, memory provenance layers, persona anchoring, pressure and drift telemetry, and reflective session logging around a locally deployed open-weight language model. By reception infrastructure, we mean runtime systems that prepare, constrain, observe, and preserve the conditions under which a conversational agent receives memory, identity-relevant context, user interaction, and corrective feedback. This differs from output containment or post-hoc moderation. We evaluate the architecture across governed and ungoverned baseline conditions, a four-condition controlled doctrine encoding experiment, adversarial sessions, and guided cultural-literacy sessions. We don't claim consciousness, sentience, or personhood. We examine whether separating authority, memory, persona, telemetry, and reflective continuity into auditable layers improves conversational stability, provenance-aware self-reference, and recovery from pressure events.