| name | derive-asset-inventory |
| description | Derive a complete target-scoped M5 asset inventory from locked canon and approved bibles. Use for requirements and semantic slots only, not specifications, generation, or processing. |
Derive an asset inventory
Scope
Translate locked world needs into one auditable inventory for one asset target.
Inputs
- Explicit
FORGE_ROOT and external WORLD_ROOT.
- Locked worldpack, target, and approved visual/audio bibles with exact hashes.
- Canonical actors, biomes, objects, actions, events, construction, UI, and audio
event catalogs.
Outputs
- One strict asset-inventory document bound to the world, target, and bible hashes.
- Requirements with ID, kind, representation, required flag, purpose, canonical
sources, and semantic slots.
- Separately identified manual additions with review rationale.
Invariants
- Preserve target dimension and delivery compatibility for mixed representations.
- Never turn an unsupported creative guess into a canon-derived requirement.
- Keep duplicate concepts, unresolved slots, and missing canonical sources as
blockers rather than silently merging or inventing them.
- Do not choose an executor, provider, prompt, extension, or production route.
Completion
Complete when every applicable world/runtime need is covered exactly once or
explicitly excluded, semantic slots are stable, manual additions are separated,
and the inventory validates. Do not write specifications or generate assets.
