This engine governs future autonomous story decisions through continuity, identity, secrets, canonical memory, cultural context and voice-fit boundaries. It extends the existing Storyworld Vault and Character Lab without writing new creative content or changing retained characters.
Persistent facts outrank generated drafts and must be checked for contradictions.
Characters remain series-scoped identities and cannot be changed by uncontrolled generation.
Character and story decisions must use the approved director lane and locale.
Future episode artifacts require consistency evidence before production readiness.
No autonomy rule creates a render or release; separate approvals remain mandatory.
Use approved cultural name entries when selecting new character identities.
Use reviewed approved voice resources before requesting any future voice generation.
Age, gender, role, tone and region fit must be resolved before a future assignment.
No single character is the center or blocker of the full platform.
Mariama remains one protected pilot character, not a global platform dependency.
No AI voice may be represented as culturally authentic without human review and consent rules.
Each selected character must belong to the selected storyworld or be explicitly introduced under review.
Retained story facts and prior episode events
Identity, biography, role and retained traits
Artifact provenance, source mode and approval lineage
Approved director lane and locale context
Known character relationships and history
Rights, QA, secure publisher and owner boundary
Hidden, foreshadowed and revealed secrets
Secrets may only be revealed according to retained narrative state and review gates.
Every autonomous episode must stay within its governed series identity and status.
Use approved voice fit when available; a gap blocks only the affected production.
Baseline mismatch detected; stop and review.
A missing authentic voice blocks only the affected production and opens a protected gap.
Voice availability, consent and permitted usage