Production Modules Overview

Operational extensions for real-world studio setups, multimodal bindings, and token dictionaries.

While the core theory and Macro-Block architecture remain universal and platform-agnostic, real-world studio productions frequently require specialized operational tooling—such as binding image assets to character names, compressing repetitive wardrobe descriptions into short tokens, or bridging multi-clip timeline continuations.

Production Modules provide these capabilities as modular, decoupled extensions. They plug into the standard 5-Block framework without cluttering or altering the underlying state mathematics.


1. Decoupled Architecture & Opt-in Model

All production modules operate under strict governance rules to prevent prompt drift and keep workflows clean:

  1. Inactive by Default: Modules are never loaded automatically. They are activated strictly upon explicit creator command (e.g., “Bind character references”, “Use Seedance syntax”, “Generate continuation script”).
  2. Platform-Agnostic Core Preservation: Operational extensions override syntax or add mapping dictionaries, but they never modify the foundational Staging vs. Execution state logic (\(S_n = f(S_{n-1} \mid \text{STAGING})\)).
  3. Strict Downstream Isolation: Reference tags (like @actor1) are bound exclusively in [INTENT]. Downstream execution blocks use canonical character names to keep timeline writing natural.

2. Production Modules Directory

Consult the dedicated documentation for each production module:

Module Name Operational Scope Activation Command Example Specification Link
Character Reference Binding Maps multimodal assets (@actor, @image, @audio) to concrete narrative personas. “Bind actor references” character_reference.md
Seedance 2.5 Syntax Overrides Replaces standard tags with native inline delimiters ({dialogue}, <sfx>) for Seedance engines. “Use Seedance 2.5 syntax” seedance25_syntax.md
Aesthetic Token Mapping Compresses long recurring lookbook definitions into shorthand tokens (LOC01, PAL01). “Use token mapping dictionary” token_mapping.md
Multi-Clip Video Continuation Handles continuity handoffs and state inheritance across successive video generation clips (@video1). “Setup video continuation pipeline” video_continuation.md

These modules ensure that as new AI video models introduce custom delimiters or reference capabilities, you can adapt your operational workflow immediately without discarding your pre-visualization discipline.

Back to top