# scene runtime
What a `.scene` folder does when it is loaded, saved, built, or swapped.
A `<name>.scene/` carries `scene.json` (the entity arrangement), an optional
`build.luau` (the code author's half), an optional `entrypoint.luau`, and a
`scene_dirty/` overlay of unsaved edits. Reading, composing and writing all
four is the scene type's own behaviour, so it lives in the type.
## Sub-modules
| Module | Responsibility |
|---|---|
| `loader.module` | Reads scene.json v6 and v7, refuses versions it does not know, composes canonical + dirty overlay, spawns the entities, and runs the entrypoint. |
| `saver.module` | Persists scene state as a delta overlay against the canonical file, promotes that overlay on save, and answers which edits are still pending. |
| `build.module` | Runs a `build.luau` inside an entity capture scope and composes what it spawned into scene records. |
| `observe.module` | The read side — the per-scene counts and costs `layers` and `loader` write. |
| `swapOrchestrator.module` | Drives multiplayer room transitions and the scene-swap gate, so a local swap does not leak despawns to peers in the old room. |
## Reaching it
```luau
local rt = require("@builtin::assetTypes.scene.shared")
rt.saver.hasPendingSceneEdits(layers.active.name)
```
Fields resolve lazily. The prelude installs `swapOrchestrator` and starts
`saver`'s dirty writer in a fixed order, which a table that required all five
on load would take away from it.
# assetType (asset type)
An asset type says what a kind of asset is: the files an asset of the kind is
made of, the code that gives it behaviour, the interface a person uses to work
with it, and how it is found, pictured and checked. A `<typename>.assetType/`
folder declares a type called `<typename>`, and every `<name>.<typename>/`
folder is an asset of it. `assetType` is itself a type, so a type is an asset
like any other: it is created, versioned, installed and contributed the same way.
## Making a type
`asset.create("assetType", "<typename>")` scaffolds a type. The type registers
the moment its folder is written, anywhere under `/zero/source/`, and from then
on `<name>.<typename>/` folders are its assets. An asset of a type that is not
registered yet is not an asset, so make the type first.
A type folder is a root: `~.x` in any file inside it names the type's own `x`,
wherever the type is installed. Keep the type's whole implementation inside its
folder and reach it that way.
What a type folder must hold is set by this type's own `type.yaml`, the same way
every type's `type.yaml` sets what its assets hold. `asset.validate` on a type
reports anything missing, and the scaffold produces a folder that passes.
## What sets your type's schema (`type.yaml`)
`type.yaml` is the structural contract every asset of the type is checked
against. The keys below are the ones the engine reads; a key not listed here is
ignored.
| Key | What it controls |
|---|---|
| `suffix` | The folder suffix of the type's assets. Defaults to `.<typename>`. |
| `primary` | The file or files that are the asset's content: what the asset resolves to when it is used. |
| `entry_files`, `primary_aliases` | Further files that load as the asset or share its identity. |
| `is_container` | Whether the suffix stays in the identities of what the asset holds. |
| `is_opaque_walker` | Whether a walk of the tree stops at the asset and registers only its primary. |
| `plural_dir`, `classify_exts` | Loose files in a folder of this name, or with these extensions, are assets of the type. |
| `is_text_default` | Whether the asset's files are text unless stated otherwise. |
| `broadcasts_subassets` | Whether assets nested inside an asset of the type are found by their own short name. |
| `root` | `true` makes every asset of the type a root, so `~.x` inside one names that asset's own `x`. |
| `contract_ref: { field, category }` | A key of the primary file that names another asset the asset is built on. |
| `files: { required, optional }` | The files an asset holds (below). |
| `allow_unlisted` | `false` reports every file the type does not declare. |
| `aliases_from: [{ file, key }]` | Further names an asset answers to, read from its own files. |
| `settings:` | Typed per-asset settings (below). |
| `inspector:` | The interface people use for the type and its assets (below). |
| `preview:` | How an asset is pictured (below). |
| `references:` | Names of other assets the type's files carry (below). |
| `indexing:` | What search reads for each asset (below). |
| `describe:` | How a missing description is generated (below). |
### The files an asset holds
Each entry under `files.required` or `files.optional` is
`{ path, kind: file | folder, one_of_group?, content? }`. `*` in a path matches
within one segment. Entries that share a `one_of_group` are satisfied by any one
of them. `content` lists checks on the file's bytes:
| Check | Passes when |
|---|---|
| `non_empty` | the file has content |
| `json_field { key, type?, non_empty?, accept? }` | the JSON holds the key (a dotted path) with that shape |
| `yaml_field { key, type?, non_empty?, accept? }` | the same for YAML |
| `described { keys?, headings?, message? }` | the file says what the asset is (see "Ask your assets to say what they are") |
Each check takes a `message`, which is what the author is told when it fails.
### Settings
`settings: { values, schema: { type | instance, export? } }` gives every asset
typed settings. `values` names the file an asset's values live in; `schema.type`
declares one schema for the whole type, and `schema.instance` names a file each
asset declares its own in. `asset.create` writes the resolved values, and a write
to the values file reaches `onChange` with what changed.
## The human side (`inspector.luau`)
A type is used by people through its Inspector, and without one the type is
usable only from code. The Inspector is the type's interface for a person: for a
procedural graph it is the graph editor, for a kit it is the tracking view of
what the kit holds and what is still missing. Writing it is part of writing the
type.
`inspector: { type, instance? }` in `type.yaml` names the files. `type` is the
interface for the type itself and for its assets; `instance` names a file an
asset may ship to add its own. Each returns:
```luau
return {
sections = { { id = "graph", title = "Graph", build = function(ctx) ... end } },
replace = { contents = { ... } }, -- replace a default section by id
order = { "graph", "contents" }, -- section order
}
```
`build(ctx)` receives the asset and builds the section. A section's `title`, and
the text its rows carry, is a string or an `i18n.text` value read from a catalog
(`i18n.text("inspectors.mesh.title")`), which follows the editor's language. `asset.create("assetType",
...)` scaffolds a working `inspector.luau` to start from. `asset.validate` on a
type that names no inspector, or names one that does not load, reports it.
## Say what your type is for (`about.yaml`)
Every type carries an `about.yaml` that says what the type is for, in three
plain-language texts:
```yaml
intent: "Sets the typeface of the text in your game: UI, labels, signs and 3D lettering."
function: "A font, from a TTF, OTF or WOFF file, made available to all of Zero's text, including 3D text."
usage: "You choose it by name for your UI or your text, and the text appears in that typeface."
```
- **`intent`** is what the type is for in someone's game, in the words a game maker uses. Up to 800 bytes.
- **`function`** is how it works today: what an asset of it is made of and what the engine does with it. Up to 1500 bytes.
- **`usage`** is what a person or an agent does with one. Up to 1200 bytes.
Write about the type as a whole, not one asset of it, and say what it does now.
Agents read these texts to learn when to reach for the type and how; people
searching find the type by them. The scaffold starts each text with
`[[describe]]` and a note on what to write; a text that still starts that way
counts as not written. `services describe` can write them.
They show up in `asset.inspect` as `detail.about`, in
`assetTypeRef:getAbout()`, at the end of the type's `types/<typename>` guide,
and at the top of the type's Inspector.
## Ask your assets to say what they are
A type can require each asset to carry a description: a file, written by the
author or generated, saying what the asset is and what it is for. Declare it with
a `described` check:
```yaml
files:
required:
- path: "README.md"
kind: "file"
content:
- { check: "described", message: "Say what this widget is for and where it goes." }
```
A file that is missing, empty or still has a line starting with `[[describe]]`
fails the check, and so does a YAML file whose named `keys` are missing or over
their `max_bytes`, or a Markdown file missing a named heading. `asset.validate`
names the asset, and a push that carries it is refused with `message`.
`services describe` finds such assets and writes their descriptions.
`describe:` steers what is generated:
```yaml
describe:
guidance: "Say what the widget shows, what input it reacts to, and where it fits in a UI."
examples:
- "<identity of an asset of this type whose description reads well>"
```
An example is an asset of your type whose description is a true, written one,
never a placeholder and never the asset being described; with none, leave
`examples` empty. A type whose assets the model cannot perceive (a sound, a
motion) says so in `guidance`, and the description is drawn only from what the
asset's own files state.
Start the file unwritten in your `template/`: a heading and one line starting
with `[[describe]]`, never example content, which every new asset would carry
verbatim. A type whose assets say what they are through their own content (a
mesh, a texture) declares no check.
## The type's code (`behavior.luau`)
`behavior.luau` returns one table. These are the keys the engine reads from it;
anything else on the table is the type's own.
### What the type's assets can do
| Key | Shape | What the engine does with it |
|---|---|---|
| `ref` | `{ [name] = function(self, ...) }` | Every function becomes a method on every asset of the type: `ref:name(...)`. |
| `modules` | `{ [name] = require(".x") }` | Code the type's assets reach with `asset.containing(__FILE__).modules.<name>`. |
| `events` | an events schema | Events an asset fires. Readers subscribe through `ref.events.<name>:connect(...)`; the type fires them. |
| `refShapes` | `{ [method] = function(self) -> (typeExpr, source?) }` | The result type of a `ref` method for one particular asset (below). |
| `namePattern` | a Lua pattern | The names `asset.create` accepts for the type's assets. Default `^[A-Za-z][A-Za-z0-9_]*$`. |
| `derive` | `{ output, module, compute, fingerprint, matches?, required?, timeout? }` | A file computed from the asset when it is committed; a `required` one holds the commit until it lands. |
Some `ref` names mean the same for every type, and the engine calls them:
- **`instantiate(self, target?, opts?) -> (root, idMap)`**: puts the asset in a scene. Defining it is what makes the type scene-instantiable. Its return is a contract (below).
- **`inspect(self) -> detail`**: what `asset.inspect` reports for the asset beyond the common record (below).
- **`preview(self, opts)`**: renders the asset's picture; it wins over the type's `preview:` recipe.
- **`previewMaterial(self) -> { material }`**: the material a `surface` preview paints.
- **`saveDefinition(self)`**: writes the asset's in-memory edits back to its files in edit mode.
Some names are answered for every asset before the type is asked, so a `ref`
entry of the same name is never reached: `getSource`, `getBytes`, `getText`,
`exists`, `deps`, `canInstantiate`, `canPreview`, `preview` (which calls yours
when you define one), `meta`, `runtime`, `events`, `typeRef`, `modules`, and the
residency flags `has_backing_asset`, `has_runtime_changes`, `cpu_resident`,
`gpu_resident`.
### When the engine calls the type
| Hook | Signature | When it runs |
|---|---|---|
| `onCreate` | `(name, opts) -> { [path] = contents }` | On `asset.create`. Returns the files of the new asset, laid over the type's `template/`. Annotate `opts`: its type is the schema `asset.create` checks the caller's arguments against. A `""` key writes a single loose file. |
| `onReplace` | `(dest, files) -> files` | On an `asset.create` over an asset already at `dest`, after `onCreate`, so what the author added can be carried into the replacement. |
| `onRegister` | `(self)` | Once per asset, when it first registers: when the world loads, on a live `asset.create`, and when an asset arrives later (installed, pulled, synced). This is what makes an asset take effect the moment it is written. |
| `onChange` | `(ref, change)` | On every write to a file inside an asset of the type. `change` is `{ path, asset, type, kind, origin }`, plus `settings` and `settingsChanged` for a write to the settings values file; `kind` is `edited` or `seeded`, `origin` is `local` or `remote`. Filter on `change.path`. Writes to one asset arrive in order, each after the hook for the one before has finished, and the hook may wait on what it reads. A write the hook makes inside its own asset arrives later like any other, so keep the hook convergent. |
| `onDependencyChange` | `(ref, change)` | When a module the asset's code requires is re-run, or a name the asset carries starts resolving. `change.kind` is `dependency`. |
| `onDelete` | `(ref)` | When an asset of the type leaves the world: deleted, or the world it is in is swapped away. |
| `validate` | `(ref) -> { { code, message, severity?, path? } }` | On `asset.validate`, after the `type.yaml` checks, for the type's own rules. `severity` defaults to `error`. Runs on every push, so a type's rules hold at publish. A hook that fails reports `validate.hook_failed`. |
### Settings hooks
| Hook | When it runs |
|---|---|
| `settingsDefaults(name, opts, asked) -> table` | Once per `asset.create`, for values the schema's defaults do not give. |
| `settingsCorrections(values, name, opts) -> (corrections?, reasons?)` | On every settings write. A correction to a field the writer named refuses the write, giving the reason. |
| `settingsDerivedDefaults(values, explicit, changed)` | On a settings write, for fields derived from others. |
| `normalizeSettings(values, stored?)` | On every read and write, to map an earlier spelling of a field to the current one. |
| `settingsHealed(values, stored)` | On a read, to repair stored values. |
| `settingsSchemaProblem(schema, ref) -> string?` | When an asset's own schema is read; a returned string refuses the schema. |
`creationInputs` (`{ [opt] = true | selectorOpt }`) names `asset.create` options
that are inputs to the type rather than settings.
### Reloading
An edit to `behavior.luau` reaches existing assets with no restart.
## How an asset is pictured (`preview:`)
| `kind` | The picture is | The type provides |
|---|---|---|
| `instantiate` | the asset placed in the preview studio through its `instantiate` | `ref.instantiate` |
| `image` | a file the asset holds, fitted to the frame | `source: { files }`, the candidate files in order |
| `surface` | the asset painted on a mesh | `mesh`, and `ref.previewMaterial` |
| `custom` | whatever `ref.preview` returns | `ref.preview` |
```yaml
preview:
kind: surface
mesh: sphere
```
`angle: [yaw, pitch]` turns the camera. The studio owns the light, the backdrop
and the framing, so an asset pictures the same in every world. A type with no
`preview:` that defines `instantiate` still answers `ref:preview()`.
A picture is rendered when something asks to view it (the editor showing the
asset, `ref:preview()`), when previews are generated on request
(`assets.generatePreviews`), and when the asset is staged. An edit alone
renders nothing. The picture records what it was made
from, so a stale one is told from a current one.
## The names your type's assets hold (`references:`)
Where your type's own format or API carries the name of another asset, declare
it here:
```yaml
references:
keys: # a top-level key of the type's files that holds a name, or a table of them
- { key: "shader", type: "shader" }
args: # an argument of a call, counting from 1
- { call: "g:node", position: 2, type: "procNode" }
fields: # a field of the options table a call takes
- { call: "fx.beam", field: "material", type: "material" }
```
`type` is the category the name resolves in. `asset.declareReferenceKey`,
`asset.declareReferenceArg` and `asset.declareReferenceField` declare the same
from a script.
## What search reads (`indexing:`)
`indexing:` is how assets are found. Search reaches an asset only through what
its type's `indexing:` names, and adds nothing of its own. That includes you: an
asset indexed badly is one you cannot find later, when you are making something
else and need it. A type whose assets are never looked for says `indexing: none`.
```yaml
indexing:
data: # text, read from the asset's files
- role: identity
extractor: verbatim
source: { file: "README.md" }
- role: interface
extractor: symbols
source: { files: ["init.luau"] }
- role: implementation
extractor: verbatim
source: { files: ["init.luau"] }
content: # the picture
- role: identity
modality: image
source: { file: "preview.png" }
facets: # what a query filters and re-ranks by
file: ".metadata"
rigged: { type: bool, from: { components: { contains: SkinnedModel } } }
roughness: { type: number, filter: range, from: { file: "mat.yaml", field: "floats.roughness" } }
```
**Roles.** Each `data` and `content` entry has a `role`, and a query matches in
two slots:
- **`identity`**: what an asset is. For anything you look at, the picture; for anything else, the text that says what it is.
- **`capability`**: what it does and how it is made. `interface` (the signatures or shape a caller uses) and `implementation` (the body) both count as capability, and a type with code usually declares both.
Ask what a person would recognise an asset of the type by, and what it does, and
point each role at the file that says it. For text, that is the asset's
description: the file your template starts with a `[[describe]]` block, which the
author writes or `services describe` fills in (see "Ask your assets to say what
they are").
**`data`.** `extractor: verbatim` embeds a file's text as it is; `symbols` embeds
the declared signatures of code. `source` is `{ file }` or `{ files: [...] }`.
Never index a file name (`source: { field: name }`): one
or two words carry nothing to match.
**`content`.** `modality: image` embeds the picture itself, so a text query
matches what the asset looks like. A type whose assets are looked at makes the
picture its identity.
**`facets`.** What a query narrows and re-ranks the matches by. `file` is embedded
whole, so keys nobody declared in it are still found. Each further key is
computed for every asset and states a value for every one, including `false`: a
file can imply "not rigged" but never say it, and a query separates rigged from
unrigged only because the unrigged ones say `rigged: false`. A key is
`{ type: bool | number | enum, filter?: range, from }`, where `from` is
`{ file, field }` (a field of one of the asset's files, dotted for nesting),
`{ metadata: key }`, or `{ components: { contains | any_outside } }` (what the
asset's entities carry).
## What blocks a push
A push checks every asset it carries against its type. These refuse it:
- a required file that is missing, a `content` check that fails, a `described` file not written yet;
- an error from the type's `validate`.
Unlisted files and warnings do not. A `described` file blocks only a push that
changes the asset. When a push is refused, the message names each asset and what
it lacks. `world.publishBlockers()` and `zm status` read the same list without
pushing.
## Typing a result from the asset (`refShapes`)
A `ref` method whose result depends on the asset (one entry per child folder, per
row of a config, per control a map declares) can only be annotated as
`{ [string]: Thing }`, which accepts any key. `refShapes` states the result for
each asset instead: a function returning a Luau type expression for that asset,
and the module whose types the expression uses. Read the asset; never run it.
```lua
-- inputMap.assetType/behavior.luau
M.refShapes = {
activate = function(self): (string, string)
local names = {}
for _, control in ipairs(M.ref.controls(self)) do
table.insert(names, control.name .. ": Handle")
end
if #names == 0 then return "", "" end
return "{ " .. table.concat(names, ", ") .. " }", "@builtin::modules.zinput.scheme"
end,
}
```
With it, `map.jump:onPressed(fn)` is checked and `map.noexisting` is reported.
A per-asset type applies where the call site names which asset: a component
field whose default names it, or an annotation `AssetRef<"inputMap", "<identity>">`.
`asset.resolve(name, category)` carries only the category, and a name computed
at runtime carries nothing, so calls on those keep the declared return. The type
belongs to the call's result: index it there, or bind what you need from it
there, and the names stay checked.
`require("@builtin::assetTypes.assetType.shared.refShapes").published()` shows
what was published: `categories` (each type's method surface) and `returns`
(per-asset results by identity, then method). An identity missing from
`returns` was never published. Results are recomputed when an asset is written;
at most 64 assets of one type are read.
## Putting an asset in a scene (`ref.instantiate`)
A caller writes one piece of code against every instantiable type, so what
`instantiate` returns is fixed:
```lua
local Instantiable = require("@builtin::assetTypes.assetType.shared.instantiable")
M.ref = {
instantiate = function(self, target, opts)
local root = Instantiable.root(self, target, opts)
-- compose the asset under `root`, now
return Instantiable.result(self, root, idMap)
end,
}
```
**In.** `target` is the entity the asset lands under, or onto for a hierarchy
type; with none, the type spawns a new root. `position`, `rotation` (three
numbers in degrees, or a quaternion), `scale`, `name` and `temporary` mean the
same for every type: `Instantiable.root` applies them to a root you make,
`Instantiable.place` to one you adopt. Document any options of your own in the
hook's `--!arg` docs.
**Out.** `(root, idMap)`. `root` is the composed root, live when you return:
build synchronously, never defer to a component's `awake`. `idMap` maps each
original id to the spawned one, so a component that composes the asset again on
every load can pass it back in and keep references into the composition. Every
call goes through `Instantiable.result`, which refuses a root that is not live
and turns a missing `idMap` into `{}`.
## Inspecting an asset (`ref.inspect`)
`asset.inspect(ref)` builds a record every asset shares (`identity`, `name`,
`guid`, `source`, `typeName`, `typeDefinitionPath`, `scope`, `origin`,
`description`, `tags`), then calls `ref.inspect(self)` for the type's own
`detail`:
```lua
M.ref = {
inspect = function(self)
-- read the asset's own files, never runtime state: inspect works on an asset
-- that has not been compiled, uploaded or registered
return { --[[ what makes this asset what it is ]] }
end,
}
```
Without it, the record is the shared part alone, and an agent inspecting the
asset learns nothing of what it is. A hook that raises leaves `detail` empty and
sets `record.warning`.
## Discovery
- `asset.list("assetType")`: every registered type.
- `asset.inspect("<typename>")`: the type, its README and its `about.yaml` texts.
- `asset.resolve("<typename>", "assetType")`: the type itself.
# Source bindings
Captures editable inputs supplied by builders, remaps their generated output ids to persistent scene ids, and retains exact source revisions for later checked edits.
```lua
local Bindings = require("@builtin::modules.sourceBindings")
local inputs, result = Bindings.capture(function()
Bindings.attach(target) -- target from SourceEditing.bind
return "built"
end)
local registry = Bindings.pack({{owner="build",targets=inputs,placements={captured="persistent"}}})
local available = Bindings.list(registry, "persistent")
local reopened = Bindings.resolve(registry, "build", available[1].id)
```
Inside a scene build, `build.source()` returns the exact `build.luau` snapshot being evaluated, including proposed source during candidate preparation. Use it with `SourceEditing.calls` to locate editable arguments, then call `build.bindSource(target)` after creating the entities named by the target. The scene captures these registrations, maps their outputs to persistent placements, and saves the registry in `authoring/state.json` under `build.bindings`. Source files discovered through bindings join the transaction read set; a changed original input or edit record refuses the build. A tool can reopen that saved registry with `Bindings.resolve` to prepare a checked source replacement.
The registry deduplicates source versions by asset guid and stores their exact-byte SHA-256 revisions. Resolving an input follows its asset identity to its current path and refuses changed bytes. Repeated registrations of one owner/input id combine known outputs only when their source occurrence and scope agree. Every output must occur in the supplied placement map. Scope completeness remains explicit. Nested capture scopes keep their registrations separate and restore the outer scope after an error.
# modules.api.engine.dirty_hot_reload
Picks up dirty-overlay changes from disk while a scene is loaded in
edit mode, applies them to the live entities, and keeps the in-memory
representation in lockstep with `scene_dirty/` files. Pairs with
`scene_saver` (writes deltas) + `scene_loader` (composes canonical +
overlay at load time).
## Parents
A record's `parent` names the entity it stands under: the entity whose id is
that text, else the entity whose display name is exactly that text. A `*` or
`?` in it is a character of the name, never a pattern.
A record whose parent no entity provides waits. An entity that does not stand
yet is not built, one that already stands keeps its state, the layer logs the
record once, and `debugState(guid).waiting` maps its id to the parent it
names. While a record waits, the scene saver leaves it as it stands rather
than writing a standing entity's older state over it, until an author edits
that entity, which the saver then records as it stands. When a record builds
an entity, the records waiting on that entity's id or display name are built
under it, so a child record that arrives before its parent's record lands
where a load of both places it. Each waiting record is built as its overlay
file states it at that moment, so a record rewritten while it waited,
including by this session's own saver, builds from its current bytes. A load
gives the same account of a record whose parent stays missing: the loader
abandons it with `parentMissing`.
# Field edits
Scene authoring uses FieldEdits to keep structured field intent separate from generated data. Each edit carries a path, the value that field was based on, and the authored value, with explicit presence flags for additions and removals.
```lua
local Fields = require("@builtin::modules.fieldEdits")
local base = {position={0,0,0}, light={intensity=1}}
local visible = {position={2,0,0}, light={intensity=1}}
local edits = Fields.diff(base, visible)
local rebuilt = Fields.compose({position={0,0,0}, light={intensity=3}}, edits)
assert(not rebuilt.blocked)
assert(rebuilt.value.position[1] == 2)
assert(rebuilt.value.light.intensity == 3)
```
- `copy(value)` validates and copies structured data without sharing mutable tables.
- `diff(before, after)` creates deterministic field edits. Objects merge by string key; arrays and asset references are atomic values.
- `compose(generated, edits)` applies authored values and reports divergence from their stored baseline. Missing parents, missing edited targets and conflicting root additions block the result. Satisfied paths identify intent the generated value already matches.
- `update(generated, previous, current, edits)` records a new authoring operation while retaining existing baselines and satisfied intent. An addition followed by its removal cancels that addition.
- `updateChanges(generated, current, changes, edits)` records original observed changes against their merged visible result. A change already satisfied by the current generated value remains an authored intent for subsequent builds.
- `keep(generated, edits, paths?)` acknowledges reviewed generated changes while retaining the authored values at the selected exact paths.
Inputs are copied, edit paths are non-overlapping dense arrays of strings, and values are finite serializable data with bounded nesting. Scene-level hierarchy, reference and component validation belongs to the scene composer consuming these field results.
# sourceEdits
Checks for authored source before it is written. `candidate` validates a
source's syntax and derives the checksum, size, kind and reference table that
storing it would state, without writing anything. `folderChecksum` derives a
folder asset's checksum from its leaves, `newIdentity` mints a UUID, and
`checkWrite` answers the canonical path a source write may take.
```luau
local sourceEdits = require("modules.api.engine.sourceEdits")
local checked = Json.decode(sourceEdits.candidate(path, text))
local guid = sourceEdits.newIdentity()
```
# playerSetupValidation
Agent-facing validation for authored player setups. A player-spawn scene is built from PlayerSpawn entities (each naming a PlayerPrototype subtree to instantiate on join) and PlayerPrototype roots (marked PrototypeOnly, carrying a Camera in their subtree and a `body` field naming the entity to adopt). `checkEntity` inspects one such entity and returns human-readable messages that name what is wrong and what to do about it; `checkScene` validates the scene's player intent against its PlayerSpawn / Camera counts; `checkActiveScene` runs both across the live active-layer scene and returns a flat `{ entity, message }` list — the queryable surface an agent reads to see what to fix.
`checkActiveScene` answers for the settled scene: it holds while a scene load or a mode-flip transition is rebuilding the live tree, then judges what the rebuild lands on. Each PlayerSpawn's prototype resolves wherever the materialisation holds it — the live entity in edit, and in play, where the loader keeps authored prototype subtrees out of the scene, the captured authored template each joining player is cloned from — so the same rules run over the same authored structure in both modes, and a prototype removed from a scene that materialises prototypes reads as the missing reference it is.
Location: `src/lua/lib/modules/api/engine/playerSetupValidation.module`
# Source operations
Require `@builtin::modules.sourceOperations` to coordinate revision-checked changes to text files in a bound world.
`capture(paths, directories?)` waits for local writes in the named scope and captures authoritative file revisions and directory membership. It includes containing asset inventories for their derived checksums. The returned snapshot is frozen and names its world and branch.
`revisionToken(snapshot)` returns a deterministic token for the guarded world, branch, file revisions and directory membership. A later capture with the same guarded revisions has the same token; guarded changes invalidate it. It leaves the snapshot and source files unchanged.
`source(snapshot, path)` reads a SourceEditing snapshot whose identity and storage checksum match the captured row. Use SourceEditing's call discovery and exact-range helpers to select the intended declaration and scope.
```lua
local Ops = require("@builtin::modules.sourceOperations")
local Source = require("@builtin::modules.sourceEditing")
local path = "/zero/source/scenes/main.scene/build.luau"
local snapshot = Ops.capture({path})
local source = Ops.source(snapshot, path)
local calls = Source.calls(source, "asset.resolve")
assert(calls[1], "select an asset.resolve declaration in this builder")
local edit = Source.replace(calls[1].arguments[1], Source.literal("props.Table"))
local plan = Ops.plan(snapshot, {{path=path, source=Source.preview(source, {edit})}})
-- Review plan.uploads and the selected declaration's affected outputs.
local prepared = Ops.prepare(plan)
assert(prepared.state == "prepared", prepared.reason)
local result = Ops.commit(plan.operation, plan.scope)
assert(result.state == "committed", result.reason)
```
`plan(snapshot, changes)` accepts disjoint text-file replacements and explicit `{path=..., remove=true}` removals. It preserves existing file identities, allocates identities for new files, derives candidate reference tables and recomputes containing asset checksums through the engine. Every mutation must be covered by the captured read set. Create asset containers through `asset.create`, then plan changes to their source files. Directory inventories participate in revision checks, including additions at previously absent paths. References are derived against the current asset registry and the captured prior reference table.
`plan` enforces ordinary source-write permissions, including computed-output protection declared by an asset type. A compiler uses `materialize(snapshot, changes)` to include its computed outputs together with their inputs. Materialization keeps the same revision, directory, identity and reference checks and respects world and library write permissions.
`prepare(plan)` first retains the exact request and its candidate pins, then uploads missing candidates under those pins. It returns prepared state once every candidate is available, already committed state after a completed operation, or failed state with the operation id retained. Repeated preparation uploads only missing candidates. Interrupted preparation is inspectable as awaitingCandidates, with the missing content hashes; resume with the retained plan or cancel it. `commit(id, scope?)` applies the retained request against its read set. A committed result acknowledges server persistence; scene activation observes the resulting files.
`inspect(id, scope?)` returns committed, prepared, awaitingCandidates, cancelled or missing state. Prepared records include the exact request, and committed records include the receipt. Use the same operation identity after a lost reply. `pending()` enumerates this identity's unfinished operations in the bound world and branch. `cancel(id, scope?)` releases a prepared operation's candidates and makes its identity terminal. Failed commits retain their prepared record for inspection and explicit resolution.
# Source editing
Tools and script builders use this module to discover Luau call arguments and edit their source declarations. A snapshot records the file identity and exact text. Every replacement uses a checked UTF-8 byte range in that snapshot; apply refuses changed files and overlapping ranges, and validates Luau or JSON syntax before writing.
```lua
local Source = require("@builtin::modules.sourceEditing")
local snapshot = Source.read("/zero/source/scenes/main.scene/build.luau")
local calls = Source.calls(snapshot, "asset.resolve")
local edit = Source.replace(calls[1].arguments[1], Source.literal("props.Chair"))
local preview = Source.preview(snapshot, { edit })
local result = Source.apply(snapshot, { edit })
assert(result.applied, result.reason)
```
Candidate evaluation uses a coroutine-local read scope:
```lua
local report, value = Source.evaluate({{snapshot=snapshot, source=plan.candidate}}, function()
return Source.read(snapshot.path).source
end)
```
Only `Source.read` uses the proposed bytes. The callback's ordinary VFS reads continue to see committed content. The report contains original read snapshots for revision guards and the candidate paths actually consumed. Nested scopes restore their caller after success or failure. Scene source replacement uses this for JSON inputs read through this module, and commits their source, rebuilt scene and refreshed bindings together. Encode JSON replacement values with `Json.encode`.
Choose the intended occurrence and scope before preparing a replacement: one call in a loop can produce many placements. Discovery reports the function spelling, original argument expressions, line numbers and zero-based byte offsets with exclusive ends. Builders supply the semantic relationship between their outputs and those inputs.
A package can retain an immutable target alongside its generated output:
```lua
local target = Source.bind(snapshot, calls[1].arguments[1], {
id = "room-west.asset",
scope = "west room declaration",
outputs = {"room-west"},
scopeComplete = true,
})
local plan = Source.prepareTarget(target, Source.literal("props.Chair"))
-- Present plan.candidate and plan.target.scope/outputs to the caller.
local result = Source.apply(plan.target.snapshot, plan.edits)
assert(result.applied, result.reason)
```
The package supplies the stable input id and output identities. A declaration
used by several placements lists that scope; use `scopeComplete = false` when
the known outputs are only part of its effect. The target retains the exact
snapshot it was bound against, so another source edit makes its apply stale.
Rebind against a fresh snapshot after resolving the intended input again.
Per-iteration changes in procedural loops need a distinct editable input or a
package transformation that creates one.
Apply returns the ordinary VFS write report. Edit-mode writes become source; a caller participating in play follows the play-shadow rules. World sync reports remote persistence and conflicts. Read a fresh snapshot to prepare the next operation.
JSON inputs use the same exact-range replacement and binding operations:
```lua
local Json = require("@builtin::modules.json")
local snapshot = Source.read("/zero/source/layout.json")
local value = Source.jsonValue(snapshot, "/rooms/1/asset")
local edit = Source.replace(value, Json.encode("props.Chair"))
local result = Source.apply(snapshot, {edit})
assert(result.applied, result.reason)
```
JSON Pointers use zero-based array indices; escape a key's tilde as `~0` and slash as `~1`. An empty pointer selects the root value. Discovery validates the complete document and refuses duplicate member names in traversed objects. Replacement preserves the surrounding bytes, including formatting. Encode JSON replacements with `Json.encode`; `Source.literal` produces Luau expressions.
An array index locates a value in the reviewed snapshot. A package resolves its stable record key, such as a room id, before selecting that index and binding its input. Changed source invalidates the prepared edit, including a reordered array.
For an array of records with string identities, select by the key:
```lua
local value = Source.jsonRecordValue(snapshot, "/rooms", "id", "east", "/asset")
local target = Source.bind(snapshot, value, {
id="east.asset", scope="east room", outputs={"east"}, scopeComplete=true,
})
local plan = Source.prepareTarget(target, Json.encode("props.Chair"))
```
Every record in the selected array must carry the named string key, and those keys must be unique. Missing identities, duplicate keys and non-record array members are refused. Resolve again against a fresh snapshot after an array reorder: the key finds the same record, while the prepared edit still checks the exact old bytes before applying.
# json
JSON encode/decode library for Luau. Encodes Lua values to JSON
strings and decodes JSON strings back to Lua values. Used for
communication with the Rust side of the engine, the VFS read/write
bridge, and any wire-format that needs JSON.
Pure Luau, no engine dependencies. Compact and pretty-printed
encoders, plus a hand-rolled decoder that streams the input by position
so it works under WASM as well as native.
## Exports
- `Json.encode(value: any, indent?: string, currentIndent?: string) -> string` — compact encode. Functions / unknown types and NaN/Inf encode as `null`.
- `Json.encodePretty(value: any, indentStr?: string) -> string` — pretty-printed encode with sorted object keys (diff-friendly).
- `Json.encodeArgs(...: any) -> string` — encode varargs as a JSON array.
- `Json.decode(str: string) -> any` — decode a JSON string. Returns the decoded value, or `nil` + error message on failure.
## Usage
```luau
local Json = require("@builtin::modules.json")
local widget = { type = "button", text = "Click Me" }
local compact = Json.encode(widget) -- '{"text":"Click Me","type":"button"}'
local pretty = Json.encodePretty(widget, " ")
local decoded = Json.decode(compact)
local v, err = Json.decode("oops") -- v = nil, err = error message
```
## Notes
- Object keys are sorted alphabetically in both encoders for consistent
output across runs.
- Numeric keys on objects are stringified at encode time (JSON has no
numeric keys). Pure-integer key sets get detected as arrays via
`isArray` and encoded with brackets.
- NaN, +Inf, -Inf encode as `null` — JSON has no representation. Round
trips through `decode` recover `null` (Lua `nil`), so they don't
preserve.
- Unicode `\uXXXX` escapes decode to UTF-8 by hand to stay WASM-safe.
Only the BMP is covered; supplementary planes via surrogate pairs
are not.
- Functions encode as `null`.
- Decode is character-streamed — no regex, no `string.match` patterns
on the whole input — so the line-and-column information needs to be
reconstructed from the position offset.
# Scene agent context
Produces bounded authoring summaries for the open scene and publishes changed summaries through the scoped world-content context channel.
`summary(status)` reduces the scene authoring status to retained-field counts, pending operation counts, conflicts, persistence state and up to three short errors. `update()` samples the root scene, caches its authoring reads by source checksum and persistence state, and publishes the result with the observed world, branch and scene identities. A rejected publication is retried on the next sample.
The summary names `sceneAuthoring.reviewRecorded` for detailed review. Values are labelled world content in agent responses.
`install()` starts one engine-lifetime sampler at quarter-second intervals, including while gameplay is paused. The prelude installs it on the engine thread. The sampler reads the current module on each sample, reports distinct failures, and retries publication. The response's `state.worldContext.values.sceneAuthoring` carries the summary under its world, branch and scene scope.
# scene_build
Turns builder code into scene records, and scene records into scene entities.
`run(builder)` opens an entity capture scope, calls `builder`, and composes every
entity it created into records — then despawns them. The builder writes ordinary
spawn code; what comes back is data. A builder that raises leaves nothing behind.
What a component the builder attached creates while running its own lifecycle
belongs to that component: the record names the COMPONENT, and the same lifecycle
runs again wherever the record is put back. That is what makes a nested build one
build — a placement the builder makes runs its own module and owns what it lands,
and the build around it describes the placement.
`reconcile(records, target, owner)` applies records to the scene under `target`,
writing `owner` and the record's place in the hierarchy onto each entity it
places as the `buildOwner` / `buildKey` attributes. A rebuild reads that pair
back off the entities under `target` and lands on the same ones, so nothing has
to be remembered between two rebuilds — the pair is in the scene, and a reload
brings it back with the entity. It returns the ids it landed on, keyed by record
identity.
A record is identified by its place in the hierarchy — the chain of names from
the build root down to it, `base/coin_3` — so dropping a line from the middle of
a builder, inserting one, or emitting two entities in the other order leaves
every other entity's id where it was. Children of one parent that share a name
carry their rank among those, `base/crate#2`, counted in the order the builder
created them.
Together these are the bake: code in, durable static-id scene content out.
## Who states an entity
An entity names the build that placed it, so `ownerOf(id)` answers which
build owns it — and answers nil for an entity an author spawned. `sourceOf(owner)`
names the file that build is written in, from the `source` the reconcile running
it passed.
That pair is what makes ownership legible where an author meets it.
`notePreview(id, values)` records what an entity carries after an author's
change and `takePreview(id)` reads it back exactly once, in the run that
states the entity again, which is where what the author set and what the build
says can be named side by side.
## What a rebuild keeps
Every entity a build places records what that build stated about it. A rebuild
whose code states a value the way the last build did leaves that value as the
scene holds it, so a hand edit to a built entity (a move, a rename, an
attribute, a component field) survives every plain rebuild. The scene's build
answers for all of it:
```lua
local report = layers.active.asset:build({ save = false })
print(report.message) -- opens with one line counting all four lists
```
`kept` lists the hand edits to entities the build placed, `driven` the fields
another component writes at runtime, `handAdded` the components and attributes
added by hand to built entities, and `handPlaced` the entities no build placed.
`layers.active.asset:build({ restate = true })` lands every value `build.luau`
states over the kept ones. `reconcile(records, target, owner, { restate = true
})` is the same at this level, and its third return value is the `kept` list.
`kept` covers every value a build states about an entity it placed: its name,
internal flag, transform, active flag, lifecycle mode, replication scope,
synced flag, attributes and components. A restate writes exactly those: a
number the scene holds within the precision a saved record keeps, and the
bookkeeping a component writes about its own runtime, stay as they are.
`drift()` reads the entity's own values alone; the build's report is the whole
answer. `guides { path: "core/scenes-as-code" }`, "What a rebuild
touches", gives the shape of every entry.
# Scene composition
Require `@builtin::assetTypes.scene.shared.composition`.
`compose(generated, intents, settingsEdits, options, previous?)` computes a candidate scene document. Each entry in `intents` is keyed by persistent entity id and contains `edits` over the value returned by `SceneRecords.saved`. An entry can also carry captured `bookkeeping` for component additions. `settingsEdits` addresses the scene document with its entities field removed.
Options supply `componentGuid(typeName)` and `componentName(guid)` resolvers. `externalEntities` explicitly identifies entities supplied by an enclosing composition.
The result contains `candidate`, `document`, `blocked`, `conflicts`, and satisfied paths per entity. Field conflicts retain authored values. Missing edited entities, invalid parent graphs, dangling declared entity references and missing component-reference targets block the candidate. A blocked result returns an independent copy of `previous` as its visible document. Its edits remain the caller's durable input.
`planMigration(canonical, visible, generated?, options)` converts a legacy saved scene and its composed dirty state into a generated baseline, entity intents and settings intents. It validates that composing these inputs preserves every normalized visible entity and scene setting. Supply a recovered build baseline where available. Without one, retained content becomes authored additions and the plan reports unresolved baseline identities. Legacy deletions remain explicit intents even when only the saved record supplies their prior values. Build identity metadata survives on retained additions. The function writes no files; the migration coordinator owns source snapshots, revision checks, recovery and activation.
The composer validates entity references declared by captured `entity_fields` annotations and component-reference envelopes. Component schema and asset-specific validation belong to the caller's semantic validation pass. The function computes data; the caller chooses when to commit and activate that data.
# Scene composition transactions
`capture(folder, inputs)` reads a scene and its authored records under a world-bound source transaction. Capture build inputs before evaluating the builder, and pass the resulting document to `rebuild`. `edit` accepts saved entity records before and after an authored change. Both compose and validate before preparing an atomic update of the accepted generated base, intent records, and `scene.json`.
`editChanges(snapshot, entities, settingsEdits, opts, retirePaths?, attribution?)` accepts compact observed field changes. Each entity row has `id`, `changes` from `FieldEdits.diff` over normalized records, and optional component `bookkeeping`. Settings use field edits over the scene fields. Changes retain their original observed values when retried against a newer snapshot. `retirePaths` names captured journal files removed atomically with the accepted composition; a newer journal revision makes the entire transaction stale. `editDocument` projects saved documents into this same field-change path.
Journal consumers pass a session/sequence receipt to `editChanges` with the corresponding captured operation path in `retirePaths`. The coordinator checks the consecutive sequence, records its checksum in accepted state and retires that source atomically. Rebuilds preserve these receipts. `assetTypes.scene.shared.journal` handles operation creation, persistence, recovery and retry checks.
`migrate` accepts a legacy visible document derived from the captured scene and overlay, together with a recovered generated base when available. It retains the canonical and visible documents as recovery history and records unresolved baseline identities.
`incorporate` prepares source replacements together with selected intent retirement. Each selection names an entity (or scene settings), the captured intent revision, and exact field paths. The candidate generated value must reproduce every selected value. Unselected intents remain. Candidate source fingerprints, the generated base, build digest and artifact advance together.
For a loaded scene’s `build.luau`, `scene:prepareSourceChange(sourcePlan, selections)` evaluates the candidate from `SourceEditing.prepareTarget`. It returns the coordinator plan together with the target’s scope and known outputs. Candidate evaluation uses isolated capture: persistent writes, external requests, deferred task work and foreign-entity changes are refused. Component lifecycle work on candidate-owned entities is captured and cleaned up.
`awaitApplied(folder, revision, scope)` waits for an accepted complete revision to arrive through the local subscription. A committed result acknowledges backend persistence; this wait establishes local visibility.
`commit` returns the durable source-operation result. Its operation identity supports inspection, retry and cancellation through `modules.sourceOperations`. Runtime scene activation observes the committed files. Structural conflicts return a plan with `ready=false` and preserve the last valid document.
# Authored scene copies
Prepare a guarded copy of an authored scene with fresh contained-asset identities, relocated JSON references and exact static Luau references, preserved authored records, and updated source bindings. Preparation leaves the destination untouched. Unsupported or ambiguous internal reference forms are refused before commit.
`prepare(source, destination)` returns a source transaction; `commit(plan)` persists it and observes the copied scene identity and artifact.
# Scene intent storage
Require `@builtin::assetTypes.scene.shared.intents`.
The store keeps one versioned JSON record per entity under a scene's `authoring/entities` folder. The filename is the entity id encoded as UTF-8 hex; the record carries its readable id, revision, field edits, per-edit attribution and component reconstruction metadata.
1. `read(sceneFolder, entityId)` captures the current record and exact source identity/bytes in a frozen snapshot. `readAll(sceneFolder)` discovers every entity record, including manual additions, and verifies that each filename matches its entity id.
2. `prepare(snapshot, generated, previous, current, bookkeeping?, attribution?)` records an operation over normalized SceneRecords values. Use the accepted generated base and the visible values before and after the operation. Independent fields merge with existing intent. Overlapping unseen changes return a conflict for the caller to retain and resolve.
3. `apply(plan)` creates an absent file with `overwrite=false`, or updates it through sourceEditing's snapshot check. A stale plan leaves the newer file intact. Empty intent records retain their revisions.
`prepareChanges(snapshot, generated, changes, bookkeeping?, attribution?)` accepts field changes produced by `FieldEdits.diff`. Each change retains its observed and proposed values. It composes those changes over the latest accepted intent, preserving independent fields and returning conflicting values for overlapping edits. Entity identity and the settings document shape are validated before a prepared result is returned. `prepare` computes these changes from its before/after values and uses the same path.
`readSettings(sceneFolder)` reads the scene-wide record at `authoring/settings.json`. Pass that snapshot to `prepare` with the scene fields after removing `entities`, then `apply` the prepared result. All three settings values must be tables. Settings carry their own revision, and entity enumeration remains scoped to entity records. Supply the returned settings edits to `SceneComposition.compose` alongside the entity intents.
Version 3 stores field history as a readable JSON array under `edits`, one entry per field edit with its earlier and later values, and exposes active dependencies through `references`. The references describe the merged visible entity, including component identities and asset envelopes; the scene asset type declares `references` as the only field of these files that carries dependencies, so an asset a historical value names stays data. Historical values remain available for conflict detection. Reads accept version 1 records and version 2 records, whose history is an encoded string; the next changed operation, and a scene copy, writes version 3. Opening a scene rewrites nothing. Snapshot records expose decoded field edits to callers. `unpackSettings(source)` and `packSettings(record)` decode and encode a settings file in that form.
Scene inspection exposes a bounded `authoring` summary, also available through `sceneRef:authoringStatus()`: recorded pending scene edits, entity intent and field counts, settings field count, and explicit read errors. `complete` indicates that these reads succeeded; the summary observes recorded state without draining queued scene writes. `hasRecordedEdits` indicates that at least one recorded change exists. These counts include retained satisfied intents.
The returned VFS report distinguishes durable source, play shadow and scratch. This per-file operation uses the world's existing synchronization and conflict protocol. Callers coordinate these records with build sources and generated artifacts.
# Observed scene operations
SceneJournal stores field edits with their original observed and proposed values. Each entry belongs to a world, branch, scene GUID and unique editing session. Session sequence numbers order operations under `scene_dirty/operations/<session>/<sequence>.json`.
`session()` creates a session token. `create(snapshot, session, sequence, previous, current, resolvers, attribution?)` normalizes the two scene documents and records their field differences. `append(sceneFolder, entry)` writes an immutable entry; it returns the path and whether an accepted receipt already settles the entry. Reusing a pending sequence with different bytes is refused. The latest retired sequence can be retried with identical bytes; older retired sequences are refused.
`createChanges(observed, session, sequence, entities, settingsEdits, attribution?)` records incremental normalized changes. `observed` carries the scene GUID, world, branch and accepted revision. The live scene writer uses this form for changed entities and settings.
`visible(sceneFolder, resolvers, session?, previous?)` composes pending entries over the locally accepted artifact. An optional session restricts which pending values are shown. Conflicts identify their journal operation. Structural conflicts retain the supplied previous complete document, or the accepted artifact. The read leaves all source files unchanged.
`resolveFields(snapshot, path, resolutions, resolvers)` chooses values for exact changed fields. Each selection has an entity id (omit it for scene settings), a field path and a `recorded` or `accepted` choice. Other proposed changes remain. Every remaining conflict must be resolved before the transaction is ready, and duplicate or absent field selections are refused.
`review(snapshot, resolvers)` returns captured operation rows, their recorded changes, current conflicts, and whether an earlier entry in the same session must settle first. `resolve(snapshot, path, choice, resolvers)` prepares a resolution of one whole reviewed operation. Choose `recorded` to retain its proposed fields against the reviewed accepted values, or `accepted` to keep those accepted values. Structural validation still applies. The resulting transaction checks the reviewed source revisions and retires only that operation with its receipt; a stale resolution leaves it intact.
`prepare(latestSnapshot, path, resolvers)` recovers the captured entry, verifies its identities and predecessor, and prepares a scene transaction. Commit it with `SceneCoordinator.commit`, then use `awaitApplied` before reading the accepted revision locally. Successful commit updates the artifact and authored intents, removes the consumed journal entry, and records its sequence and checksum atomically. A stale operation, validation failure or field conflict retains the entry for inspection and resolution. The accepted scene state retains one receipt per editing session.
# `scene.shared.loader`
Luau-side scene loader. Reads `scene.json` v6, dispatches to engine
FFIs, runs the sibling `entrypoint.luau`.
## Public surface
```lua
local loader = require("@builtin::assetTypes.scene.shared.loader")
loader.load(ref, opts?)
-- ref: AssetRef<scene> | identity string | VFS path
-- opts.layer: layer name (default "main")
-- Returns: the Scene proxy (with player_config + camera_config attached).
loader.unload(layerName)
-- Wrapper around layers.unload(name); symmetric to M.load.
-- Cannot unload "main".
```
## Behavior
1. Resolve `ref` to a VFS path; append `/scene.json` if needed.
2. `vfs.read` + `Json.decode` (from `modules.json`) → body table.
3. Version check — refuse anything but `version: 6`.
4. Attach `body.player` + `body.camera` to the Scene proxy as
`player_config` / `camera_config` (consumed by player_spawner +
camera_spawner).
5. Activate the target layer via `layers.find(layerName)` (the loader
threads the layer through `entity.spawn` opts), spawn `body.entities`
in two passes (shells + transforms first, parents + components
second), then restore the previous active layer.
6. Apply `body.lighting` via `lights.setup` + `lights.addPointLight`.
Field name mapping: `directional` → `sun`, `clear_color` →
`clearColor` (both engine API names and v6 field names are accepted).
7. Auto-discover sibling `entrypoint.luau` via `vfs.exists`; if
present, compile with `loadstring` + run.
## Performance
~5 ms for a 1000-entity scene (≈5k FFI calls × ~1 µs each). Acceptable
for a load-time operation. If a particular call class becomes hot,
batch primitives can be added in a narrow follow-up.
## Why pure Luau
Per § 10 of the unification integration design, the scene format is
data + dispatch — no Rust-only primitives needed. The existing ECS
FFI (`entity.spawn`, `entity(id).component.add`, etc.) covers
everything. Keeping the loader in Luau means iteration speed,
auditability, and hot-reload.
## FFI corrections vs. the original plan
The plan was drafted against assumed API names. The actual engine FFIs differ:
| Plan assumed | Actual engine API |
|---|---|
| `json.parse(s)` | `require("..json").decode(s)` (`modules.json`) |
| `json.encode(t)` | `require("..json").encode(t)` |
| `entity.spawn(n, { layer }).id` | layer activation via `layers.find(layer)` + `entity.spawn(n, opts).id` |
| `entity.setParent(id, pid)` | `entity(id).setParent(pid)` |
| `unloadLayer(name)` | `layers.unload(name)` |
| `lights.setup({ directional })` | `lights.setup({ sun })` |
| `lights.setup({ clear_color })` | `lights.setup({ clearColor })` |
| `lights.addPointLight(table)` | `lights.addPointLight(name, x, y, z, opts)` |
The loader accepts both the v6 field names and the engine API names for
the lighting fields so that round-tripping through the scene saver and
back works without a conversion step.
# scene_observe
The record of what a scene load did, and what each loaded scene costs. Backs
`layers.observe`, `layers.lastLoad`, `layers.lastUnload`, `layers.loadHistory`,
`layers.problems`, `layers.whyPartial`, `layers.inventory`, `layers.cost` and
`layers.resetCostWindow`, and the `scene` toolbox's `observe`, `whyPartial` and
`cost` above them. `layers` and `scene_loader` are its writers.
## One record per load
A load opens a record; the loader, the lifecycle dispatchers and the scene's
own build write their failures and phase timings into it; the load's own
completion point closes it with the duration the engine already measured for
its `DONE` line. Reading is lazy — nothing here walks the world, and the only
writer that runs every frame is `noteUpdate`, which the entrypoint tick calls
once per scene that declares one.
```lua
local r = layers.lastLoad()
print(r.name, r.outcome, r.durationMs)
print(r.entities.added, "arrived,", r.entities.removed, "left")
print(r.phases.teardown, r.phases.instantiate, r.phases.settle)
```
`outcome` is `"ok"` when everything the scene declared was produced,
`"partial"` when the load finished with failures in it, `"failed"` when the
load raised and left no layer, and `"unchanged"` when the scene asked for was
already the active root — that load rebuilt nothing, so the layer keeps the
record of the load that built it and still answers why it is not whole.
`reason` names the nearest cause from the closed
set — `loaderRaised`, `entrypointCompileFailed`, `entrypointBodyRaised`,
`entrypointRaised`, `buildRaised`, `entityFailed`, `parentMissing`,
`parentRefused`, `parentAbandoned`, `componentUnresolved`, `componentRefused`,
`subscriberRaised`, `updateRaised` — ranked so it names the thing to fix rather
than the last thing to break.
## Failures reach the layer they belong to
Every write of a failure calls the subscriber `layers` registers through
`M.onFailure`, which republishes the record onto that layer's proxy: `ok`
false, `failures` the records, and `"partial"` in place of `"ready"` once the
layer has settled. A tick that starts raising on frame 400 moves the layer the
same way one raised during the load does.
A failure identical to one already held raises that one's `count`, so a tick
raising every frame keeps one record. `FAILURE_LIMIT` bounds the distinct
records a report holds, and `failuresOmitted` states how many arrived past it.
## Cost
`noteUpdate` accumulates each layer's entrypoint tick into a per-guid entry —
calls, total, last, max, average and how many raised — summed across the window
`M.window()` reports. `M.resetWindow()` opens a new one, leaving the load
history alone. An unload drops the layer's entry.
# Scene record projection
Require `@builtin::assetTypes.scene.shared.records`.
`saved(record, componentGuid)` converts a persisted entity into `{value, bookkeeping}`. `captured(record, ids, componentGuid)` converts an entity-record capture and remaps captured entity references to persistent placement ids. The resolver receives a component type spelling and returns its stable identity; unresolved types must raise an error. Captured entities require an entry in `ids`.
`value` holds editable fields with normalized transform and lifecycle defaults. Components are grouped by stable type identity: `named[name]` addresses one named instance, while `anonymous` preserves an ordered group. Saved `instance_index` markers are reconstruction metadata, and restoration emits them alongside named `instance_name` markers. FieldEdits treats that anonymous array atomically. Asset references compare by guid and type. Component references retain their entity, instance information and stable component type.
`bookkeeping` holds build ownership attributes, component order, and captured reference paths and row-field declarations. These capture annotations describe reconstruction and stay outside field comparisons. Compose edits over `value`, then call `restore({value=composed, bookkeeping=current.bookkeeping}, componentName)` to produce the saved record. The reverse resolver supplies the current component type spelling. Surviving instances retain their order; added instances are appended in deterministic order.
The module validates structured input before copying it. Duplicate named instances and missing identities fail before any scene mutation.
# `scene.shared.saver`
Luau-side scene saver. Writes `scene.json` v6 — **authored intent
only**. No captured runtime state.
## Public surface
```lua
local saver = require("@builtin::assetTypes.scene.shared.saver")
saver.save(name, opts?)
-- name: logical scene name ("main") OR full VFS path
-- opts.layer: layer name to read from (default layers.active.name or "main")
-- Returns: the absolute VFS path written.
```
## What is captured
| Field | Source | Notes |
|---|---|---|
| `format`, `version` | constants | Always `"scene"` + `6`. |
| `entities` | `entity.findAll()` walk | Excludes `temporary` entities + the spawner-managed Player + primary Camera. |
| `entities[].transform` | local position / rotation / scale | NOT world transform — local only. |
| `entities[].components` | component.list + component.get | Strips `_`-prefixed runtime fields. |
| `entities[].attributes` | `attribute.list` + `attribute.get` | The entity's key/value bag, written as a `{ [key] = value }` map; omitted for an entity that holds none. Holds the authored keys — a `_`-prefixed key is the engine's own session bookkeeping. |
| `lighting` | `lights.snapshot()`, filtered | Scene-level `clear_color` + `sky` only. Directional / ambient / point lights persist as `Light` entities in `entities[]`. |
| `player` | `sceneProxy.player_config` | Carried forward from load-time, authored only. |
| `camera` | `sceneProxy.camera_config` | Carried forward from load-time, authored only. |
## What is NOT captured
- Local player's position / rotation. Runtime drift; not authored intent.
- Primary camera's transform. Runtime; behavior-driven.
- Any entity flagged `temporary = true`. Debug helpers, runtime markers.
- Internal component fields prefixed with `_` (e.g. `_velocity` on CharacterController).
The spawner-managed entities (Player + primary Camera) are excluded
because they're created on load by player_spawner / camera_spawner
from the scene's declarative `player` + `camera` blocks. Saving them
would double them on the next load.
## Parts
The module is a folder of parts, each a module of its own that `init.luau` requires:
- `scene_saver.serialize`: one live entity as the record a scene file carries.
- `scene_saver.recordDiff`: the comparisons the record diffs are built from.
## When to use
`layers.active:save(name)` (Phase 6) routes through this module. Callers
that want a state snapshot (save-game systems, undo stacks) should use
`world.debugSaveToDisk(...)` instead — that captures runtime state by design.
# @builtin::assetTypes.scene.shared.swapOrchestrator
Owns the user-scope orchestration for scene swaps: leaves the prior
scene's multiplayer room, gates concurrent loads via the
`__scene_load.begin/finish` token, drives `scene_loader.M.load` for
the new scene, and joins the new scene's room. Keeps the swap atomic
from a multiplayer-sync standpoint (room transition + entity churn
happen inside a single barrier).
# Module (asset type)
A module is a reusable, stateless (or globally-stateful) Luau library
that other code pulls in with `require()`. Modules are the unit of
factored-out logic in a world: helpers, math libraries, format
parsers, registries, shared algorithms.
## When to use one
- You have logic you want to share across multiple components,
services, scenes, or tools.
- You have a registry, lookup table, or shared cache that callers all
read from / write to.
- You want a stable API surface (`M.fn`) other code can depend on.
If the logic is bound to one entity, use a `.component`. If you want a
named UI of fields rather than functions, use a `.preset`.
## Where it lives
- Source: `/zero/source/.../<name>.module/`
- Identity: the source path with `/` written as `.` and each segment's
type suffix dropped — `/zero/source/demo/beats.module` is
`demo.beats`, and `/zero/source/kit.package/beats.module` is
`kit.beats`, the container's name kept and its `.package` suffix
gone. `asset.create` returns this string, `asset.resolve(...).identity`
reads it back, and the `require` forms below write it as
`<identity>`.
- Folder shape:
- `init.luau` (or `init.lua`) — module body. **Required.** Returns
a table (conventionally named `M`).
- `README.md` — instance-level documentation. **Required.**
- `.metadata` — agent-editable tags + free-form fields. **Required.**
## How to create one
```luau
asset.create("module", "<name>")
-- Creates: /zero/source/<name>.module/
-- init.luau (canonical `local M = {} … return M`)
-- README.md (instance README template)
-- `folder` places it in a subfolder of /source, and that subfolder
-- leads the identity: `{ folder = "lib" }` produces `lib.<name>`.
-- `into` authors it inside a resolved container (a package, a toolbox),
-- whose own identity leads the same way:
asset.create("module", "<name>", { folder = "lib" })
asset.create("module", "<name>", { into = asset.resolve("myPkg.package") })
```
## How it operates
1. **Registration.** Writing `init.luau` into a `.module/` folder
indexes the asset and registers the module path with the resolver.
2. **Loading.** The first `require("<identity>")` runs
the file's body and caches the returned table. Subsequent requires
from anywhere return the **same table** — this is the
shared-state lever for "module-local" registries. The body runs in
an environment of the module's own: a name it declares without
`local` is readable by the module's own functions and by no other
script, so an export reaches its callers through the returned table
and nothing else. Requiring a component's folder runs its chunk the
same way, once, and its `update`, `start` and `public` stay in that
environment; a component runs on an entity through
`entity(id).component.add(name)`.
3. **Hot reload.** Editing the file re-runs the module body and folds
what it returns into the table already loaded, so that table keeps
its identity: a caller that captured it with `require` reads the new
source's functions and values through the reference it already
holds, and a later `require` returns that same table. A key the new
source drops goes away; a key something set at runtime stays. The
body's own `local`s start again from their initial values, so state
that must outlive an edit lives in `modules.state()`
(`guides { path: "core/scripting-and-tasks" }`).
4. **Identity resolution.** A module is reached by its `<identity>`,
and where the caller lives decides which spelling of it resolves.
For a module in a **world**, from a caller in that same world —
another module, a component, a scene entrypoint, or an `execute`
chunk:
- `require("<identity>")` — the identity on its own, resolved
against the caller's own root. It carries every folder segment,
so a module created with `{ folder = "demo" }` is
`require("demo.beats")`.
- `require("@root::<identity>")` — the same root-relative
resolution, written out.
- `require(".sibling")` — a module in the same folder, by its bare
name; `..name` steps up one folder, `...name` two.
Logs, stack traces and hot-reload notices name that same module
`@local.source.<identity>` — its key under the world source root.
The `<identity>` tail of that key is the form above, so a trace
reading `@local.source.demo.beats` is `require("demo.beats")`.
For a module in a **library** — content under
`/zero/source/libs/@<lib>/` — `@<lib>::` addresses that library's
root from anywhere: `require("@builtin::modules.transform")`.
Mounting a world as a library moves its files under such a root and
flips the caller root to `@<lib>::`, which is what keeps the
root-relative forms above resolving across the move.
## Discovery
- `asset.list("module")` — every registered module.
- `asset.inspect("<name>")` — public functions, source path, this
type README.
- `ls /zero/source/<name>.module` lists the files the folder holds.
## Authoring conventions
- Return a single `M` table from `init.luau`. Top-level statements
with side effects run on first require — useful for one-time
initialization, dangerous if they touch the engine before it's
ready.
- Write exported functions as `typed function`, not `function`: the
argument types are enforced at the call and the checker knows the
signature, so a caller's mistake is reported where they made it.
- Annotate exported functions with `--!desc` / `--!arg` / `--!return`
/ `--!example` so the LSP, `tools.list`, and `asset.inspect`
surface them. A function with no `--!desc` is reachable only by
someone already reading this file.
- Declare a VALUE with `Field.<kind>(default, mode, description)`, and
a table whose members are reached through a metatable with
`--!members <TypeName>` — the latter gives the surface its accepted
member set, its type, and its documentation from one table. See
`guides { path: "core/authoring" }` for all four rules and
`modules/api/engine/entity.module/members.module` for a worked
member table.
- Keep module state in module-local upvalues or in `modules.state()`.
A name the body declares without `local` lives in the module's
environment, which every run of the module shares: a function the
previous run left in place reads what the reload assigned. Module-locals start again from their
initial values with each hot reload; the table `modules.state()`
returns is the same table on both sides of every reload.
- Prefer focused modules over kitchen-sink modules. If two halves of a
module have no shared state, split them.
## Common pitfalls
- **Cyclic requires.** A `require` chain that loops will return the
partial table (the half built before the cycle was detected).
Design for one-way dependencies.
- **Engine-time side effects.** Don't `entity.spawn` at module
top-level — the engine may not be ready. Expose an `M.init` and
call it from a scene entrypoint.
- **Hot-reload + module-local cache.** A module that caches
expensive computation in upvalues loses that cache on reload —
fine for dev, but be aware in performance work.
- **`init.lua` vs `init.luau`.** Either works, `.luau` is canonical.
- **Writing one while play runs.** A `/source` write made while play is
running lands on the play shadow, and a guarded play-exit discards it
unless it is kept. One `.module` write shadows `init.luau`, the
`README.md` and the `.metadata` together.
`vfs.write(path, bytes, { durable = true })` skips the shadow, and
`vfs.promotePlayShadow` keeps what is already on it — the `core/vfs`
guide has the model and what each route costs.
## Related types
- `.component` — for entity-bound state + lifecycle hooks.
- `.service` — generates content you don't have yet (a mesh, sound, texture, …) via a metered provider.
- `.tool` — for a single agent-callable function with a YAML schema.
- `.package` — to group several modules + components + scenes into
one shippable folder.
## In the Inspector
Exports: what the module exports (name, kind, signature), parsed from its source without running it, with **Check** (the LSP's diagnostics for its entry file) and **Open**.
# Model
Loads and displays a 3D mesh on an entity. Disabling the component hides the mesh. Supports procedural meshes (`cube`, `sphere`, ...), model files, and remote URLs. Tint and outline are shader effects applied to the mesh.
**Material lives on this component.** There is no standalone `Material` component — a material only renders where there is a mesh to render it on, so the `material` field is part of Model (and SkinnedModel). Set it at add time or assign the field later; properties are registry-wide.
Public fields: `model`, `material` (`AssetRef<material>`), `tintR/G/B`, `tintBlend`, `outlineR/G/B`, `outlineIntensity`, `outlineOccluded` (`"hide"`, the default, hides the rim where geometry stands in front of the mesh; `"dashed"` draws it there in dashes, the `occluded` choice of `@builtin::modules.outline`).
Methods: `:setTint(color, blend?)` (color: `{r, g, b}` array or `{r=, g=, b=}` map), `:clearTint()`, `:setOutline(color, intensity?)`, `:clearOutline()`, `:setMaterialProperty(prop, value)`, `:getMaterialProperty(prop)`, `:getMaterialPropertyNames()`.
```luau
entity(id).component.add("Model", { model = "cube", material = "gold" })
entity(id).component.get("Model").material = "checkerboard" -- swap material
entity(id).component.get("Model"):setTint({1, 0, 0}, 0.5)
entity(id).component.get("Model"):setMaterialProperty("roughness", 0.2)
```
**Tint is synced.** Every `tint*` field replicates, so `:setTint` is the colour every participant sees. A colour that changes every frame on many entities (a stress map, a heat map, a damage flash, per instance colour) belongs in an instance-data lane instead: wear `@builtin::materials.instance_color` (or `instance_color_unlit`) and write `renderer.instanceData.set(id, 2, r, g, b, coverage)`, which replicates nothing and costs about a twentieth of a tint write. `guides { path = "topics/rendering-instance-data" }` has the lane layout.
# PlayerPrototype
Marks the root of a player-setup subtree — the entity and children that define what a joining user becomes (avatar slot, spawn behaviour, controller). On awake it flips the entity to PrototypeOnly participation, so the prototype lives in the editor and drives spawning without being saved as a world entity itself; `kind` and `label` describe it for authoring surfaces. In edit mode it draws a cyan silhouette outline so the prototype root is easy to pick out, and surfaces its own setup-validation messages (missing camera, avatar slot, or wrong participation) on the component's error channel.
Location: `src/lua/lib/components/PlayerPrototype.component`
# BoxCollider (component)
A box collision shape. Works with or without a `Physics` component: on its own
it is static collision geometry, and alongside one it moves with the body.
## Sizing
`half` is the box's half-extents in the entity's local space, multiplied by the
entity's world scale. Leave it unset and the box derives from the entity's model
bounds, so a bare `BoxCollider` matches what is rendered at any scale.
```lua
entity.find("crate").component.add("BoxCollider", {}) -- fits the mesh
entity.find("wall").component.add("BoxCollider", { half = { 2, 4, 0.2 } }) -- authored
```
## Fields
| Field | Meaning |
|-------|---------|
| `half` | Half-extents. Unset derives from model bounds. |
| `isTrigger` | Report overlaps without blocking. |
| `friction` | Surface friction. |
| `restitution` | Bounciness. |
| `rollingResistance` | How strongly the surface resists a body rolling on it: a rolling body slows at this many times gravity on level ground, so a ball rolled at v stops after v^2 / (2 * rollingResistance * 9.81). 0 rolls on. |
| `spinningResistance` | The same for a body spinning on the surface about the contact normal. 0 spins on. |
| `surface` | The surface the collider is made of, by name: one `Physics.defineSurface` defined gives it its friction, restitution and resistances in place of the fields above, and contacts, contact events and ray and shape casts report the name either way. Empty names none. |
| `oneWay` | The direction, in the collider's own frame, bodies pass through it along: a body moving that way goes through, and a body meeting it from the side the direction points to stands on it (a jump-through platform is `{0, 1, 0}`). `Physics.passThrough(entity)` drops a standing body through. Unset blocks from every side. |
| `surfaceVelocity` | The velocity, m/s in the collider's own frame, its surface moves at along itself while the collider stays put: friction carries what rests on it (a conveyor belt, a treadmill, an escalator). Write it at runtime to change the speed or direction. Unset holds the surface still. |
| `contactTag` | The tag the world's contact filter reads this collider by: `Physics.setContactFilter(tagA, tagB, mode)` makes two tags pass through each other (`"ignore"`), report contacts and push nothing (`"sensor"`) or collide (`"solid"`). Empty names none. |
## Composing
A collider on a child entity attaches to the nearest ancestor carrying a
`Physics` component, so a body made of several shapes puts each on its own
entity and each keeps its own material and trigger flag.
Past a few dozen boxes sharing one material, `CompoundCollider` is cheaper: it
is a single broadphase volume where separate components are one each.
# Physics
Adds rigid body dynamics to an entity. Body kind is `"dynamic"` (affected by forces), `"static"` (immovable), `"kinematic"` (placed by script at the pose its transform holds) or `"kinematicVelocity"` (carried by the solver at the velocity script gives it with `Physics.setVelocity`). Both kinematic kinds push dynamic bodies and are not pushed back.
Public fields: `kind`, `gravityScale`, `mass`, `linearDamping`, `angularDamping`, `ccd`, `interpolation`, `lockRotationX/Y/Z`, `lockTranslationX/Y/Z`, `dominance`, `canSleep`, `sleepLinearThreshold`, `sleepAngularThreshold`, `centerOfMass`, `inertia`, `softCcd`, `extraSolverIterations`, `extraPgsIterations`. Every field reaches the solver on the next step when written, and `Physics.bodyState(id)` reads each back as the solver holds it. A value outside what a field takes is refused with an error naming the field.
`dominance` (-127 to 127) decides who pushes whom between dynamic bodies: a body in a higher group pushes one in a lower group and is not pushed back, the way a player character shoves crates without slowing. Bodies in one group push each other as usual.
`canSleep = false` keeps the body simulated every step however slowly it moves (a probe coasting in zero gravity, a vehicle idling at rest); `sleepLinearThreshold` (m/s at the body's farthest point, default 0.05) and `sleepAngularThreshold` (rad/s, default 0.5, for a body with no colliders) set how still it must stay for half a second before it sleeps. `Physics.sleep(id)` puts a body to sleep at once.
`centerOfMass` (local, metres) and `inertia` (principal moments, kg·m²) replace the values the colliders give the body; leave them unset to derive them. A car body with `centerOfMass = { 0, -0.5, 0 }` resists the roll that tips it with the derived centre.
`softCcd` (metres) predicts contacts that far ahead of the body each step, which catches a fast body before it passes through a thin wall at a fraction of the cost of `ccd`. Set it to about the distance the body travels in one step (speed / tick rate).
`extraSolverIterations` and `extraPgsIterations` (0 to 16 each) stiffen one structure without raising the solver world-wide: the island the body is part of (everything jointed to it or touching it) runs the world's `solverIterations` plus the largest extra substeps any of its bodies asks for, and the world's `pgsIterations` plus the largest extra passes. A body held under a heavy load by impulse joints (a light link of a chain holding a heavy block) also asks its island for substeps short enough to hold it stable, which `Physics.bodyState(id).solvedSubsteps` reads. Put the same values on every member of a bridge, a chain or a ragdoll; the island's step costs about that many more substeps, and the rest of the world costs what it did.
`ccd` (continuous collision detection) sweeps the body along its motion so it cannot pass through thin geometry between two steps. It is off by default: the sweep runs on every body that travels further than its own thickness in one step — most of a collapsing structure or a debris burst — and it dominates the physics budget when a scene has many such bodies. Turn it on for the individual bodies that would visibly tunnel: bullets, thrown props, a wrecking ball.
```luau
entity(id).component.add("Physics", { kind = "dynamic", ccd = true })
```
A body on an entity under a parent lives in world space. It starts where its entity stands in the world, with the parent's position, rotation and scale applied, and a write to the entity's `position`, `localPosition`, `rotation` or `localRotation` places the body where that write puts the entity, in the same call: a ray cast right after the write meets it there. The one exception is a `kinematic` body while the clock runs: the write aims it, and the next step sweeps it there, carrying the dynamic bodies in its path. While the clock runs, the solver moves a `dynamic` or `kinematicVelocity` body and its parent does not carry it: moving or turning the parent leaves the body where the solver has it, and each step writes the entity's local transform from the body's pose, so the entity stays on the body. A `static` or `kinematic` body goes where its entity goes, its parent's moves included. With the clock held (edit mode, a paused session), every body goes where its entity goes. A change of parent that keeps the entity's local pose (`setParent` or `unparent` with `keepWorldTransform = false`) is a write of that kind: the entity's local pose read under its new parent (or as world, for `unparent`) places its body, every kind of body, whether the clock runs or is held, as a write to its `localPosition` does. The placement stops a `dynamic` body and leaves a `kinematicVelocity` body the velocity it was given; a `kinematic` body while the clock runs is aimed there, as for any write. `keepWorldTransform = true` leaves the entity where it stands in the world, and its body where it is, at the velocity it has. Colliders on the entity's children go with its body.
`mass` is the body's total mass in kilograms. At its default of 0 the mass comes from the space the body's solid colliders fill, at 1 kg per cubic metre, so a bigger collider makes a heavier body; a value greater than zero sets the total mass exactly. Either way the mass is spread evenly over that space: where two colliders overlap, the space they share is counted once, so a ball inside a box weighs what the box weighs, and a trigger (`isTrigger = true`) fills no space and adds no mass, inertia or shift of the centre of mass.
A body's colliders are the ones on its entity and on every entity beneath it that carries no `Physics` of its own, whatever order the components were added in: a collider added to a child before its parent got `Physics`, or a child moved under the body later, joins the body, and a child moved out from under it, or whose body's `Physics` is removed, stands alone as static geometry. The body's mass, centre of mass and inertia come from the shapes, poses and sizes of those colliders (a `mass` above zero is spread over them), so a car built from a chassis and child parts rolls and pitches the way its parts are laid out. `Physics.bodyState(id).colliders` lists them.
A body whose entity holds no collider is given a stand-in box: one fitted to the entity's model, or a 1 m box (half extents 0.5) when it renders none. It collides with everything and `Physics.bodyState(id).colliders` lists it. A collider component added to the entity replaces it, and the first solid collider (one that is no trigger) on an entity beneath the body takes its place as it joins the body, in the same frame when both are added together. A body whose colliders beneath it are still waiting for their mesh, or sit on an inactive entity, rests on the stand-in until one of them is built. The stand-in is given as the body is built: a built body whose last collider is removed or deactivated, or moves out from under it, holds no collider until another one joins it.
Methods: `:applyForce(x, y, z)`, `:applyImpulse(x, y, z)`, `:applyTorque(x, y, z)`, `:setVelocity(x, y, z)`, `:addVelocity(dx, dy, dz)`.
```luau
entity(id).component.add("Physics", { kind = "dynamic" })
entity(id).component.get("Physics"):applyImpulse(5, 0, 0)
```
# Light
Adds a light source to an entity. The entity's world transform determines the light's position (point) or direction (directional). Point light positions are auto-synced from that world transform, so a light on a child entity burns where the entity stands — the position `entity.position` reports. A point light's `intensity` is on one scale with the `SpotLight` component's `intensity`/`brightness`: the same number at the same `radius`/`range` puts the same light on a surface either kind faces from the same place, and a spot spends it on the cone it opens on rather than all around itself.
Kinds: `"point"`, `"spot"`, `"directional"` (the scene's sun), `"ambient"`, `"distant"` (a parallel light beside the sun).
Public fields: `kind`, `colorR/G/B`, `intensity`, `radius`, `directionX/Y/Z`, `castsShadows`, `lightChannels`, `mobility`. `range`, `color` and `direction` are aliases accepting the composite/renamed forms.
Methods: `:setColor(color)` (color: `{r, g, b}` array or `{r=, g=, b=}` map), `:setIntensity(i)`, `:setRadius(r)`, `:setDirection(dir)`, `:setKind(lightKind)`.
`kind = "spot"` opens a cone along the entity's forward axis, at the engine's default 30° outer and 20° inner half-angles; the `SpotLight` component is the one that carries the cone angles and the `face` axis as fields. `"directional"` and `"distant"` are parallel lights: they arrive from the same direction at every point in the world and no distance attenuates them, so their `intensity` reads against the sun's rather than against a point light's. The `SpotLight` README carries the rest of how to balance a mixed point-and-spot rig.
```luau
entity(id).component.add("Light", { kind = "point", intensity = 2, radius = 10 })
entity(id).component.add("Light", { kind = "directional", direction = {-0.5, -1, -0.3} })
```
# ProceduralSky
The editable atmospheric sky: a day/night gradient with a sun disc, glow, stars
and a moon, all derived procedurally from the directional light's elevation. Add
it to an entity like a light:
```lua
entity.spawn("sky").component.add("ProceduralSky", { timeOfDay = 18.5 })
entity.spawn("sky").component.add("ProceduralSky", { preset = "sunset" })
entity.spawn("sky").component.add("ProceduralSky", {
zenithColor = { r = 0.05, g = 0.1, b = 0.3 },
horizonColor = { r = 0.9, g = 0.4, b = 0.2 },
})
```
A `ProceduralSky` owns its material outright: it registers a **runtime GPU
material** of its own over the builtin procedural-sky shader and pushes every
visual parameter (colours, sun, stars, moon, turbidity, exposure) straight to
that record. The component's fields **are** the material definition — no
`.material` asset backs it, so retuning a scene's sky changes that scene's sky
and nothing else. The renderer reads the values from the material's group(1)
uniform like any other material — there is no sky-specific render path.
Those values also travel the native `Sky` bridge into the scene sky config,
alongside the time-of-day / auto-cycle / sun-sync controls. That config is what
`sky.get()` reports and what a saved scene records, so both name the sky being
drawn.
With `syncSunToLight` on (the default), the sky's `timeOfDay` is what the scene
is lit by: the scene's directional light takes its direction, colour and
intensity from the sun's position, so the sun in the sky and the sun the scene
is lit by are the same sun through a whole day/night cycle. Turn it off for a
light the sky leaves alone.
The intensity that arrives is the day/night curve between night and noon, and
`sunPeakIntensity` is the noon end of it — `1.0`, the sky's own daylight, unless
a scene asks for more. A harder sun on water or snow is stated here; setting the
light itself does not survive, because the sky writes over it every time the
hour moves.
Like the directional light, the sky is conceptually **singular per scene**. A
`ProceduralSky` fully defines its material on `awake` (every parameter from its
own fields), so switching scenes never inherits a previous scene's overrides.
Day/night comes from the sun's elevation, so evening presets render dark —
drive the look by `timeOfDay` (which orients the sun).
## Fields
| Field | Type | Default | Meaning |
|---|---|---|---|
| `preset` | string | `""` | Named look applied before explicit fields: `clear_day`, `sunset`, `sunrise`, `overcast`, `night`. |
| `timeOfDay` | number | `14.0` | 0..24 hours; orients the sun and the day/night gradient. |
| `autoCycle` | bool | `false` | Advance `timeOfDay` each frame. |
| `cycleSpeed` | number | `60.0` | Game-seconds per real-second for the auto cycle. |
| `syncSunToLight` | bool | `true` | Drive the directional light from the sun. |
| `sunPeakIntensity` | number | `1.0` | What the sun's light measures at noon; the day/night curve runs from night up to this. |
| `zenithColor` / `horizonColor` / `groundColor` | color | — | Sky gradient colours. |
| `sunSize` / `sunIntensity` | number | `0.02` / `20.0` | Sun disc size + brightness. |
| `starsIntensity` | number | `1.0` | Night star-field intensity. |
| `starsTwinkle` | number | `0.0` | How far each star's brightness swings over time; `0` holds the field still. The `night` preset twinkles at `0.3`. A twinkling sky changes the picture on every frame, so the renderer keeps drawing under it. |
| `moonSize` | number | `0.03` | Moon disc size. |
| `turbidity` | number | `4.0` | Atmospheric haziness. |
| `exposure` | number | `1.0` | Sky exposure multiplier. |
## Methods
| Method | Description |
|---|---|
| `sky:setTimeOfDay(hours)` | Set the time of day (0..24). |
| `sky:applyPreset(name)` | Apply a named look. |
| `sky:setZenithColor(c)` / `setHorizonColor(c)` / `setGroundColor(c)` | Set a gradient colour (`{r,g,b}` array or map; >1 auto-scales /255). |
# Camera
Manages viewport priority, render-to-texture, and capture. State is stored in the native `Camera` ECS component; the Rust camera system handles render scheduling and render targets.
Public fields: `fov`, `near`, `far`, `priority`, `textureHandle` (the guid of the texture the camera renders into; empty = main viewport), `renderLayers` (which render layers this camera draws — a space-separated spec of names, e.g. `"all"`, `"all !ui"`, or `"default sky"`; `ui`/`sky`/`debug`/`EditorUI` are built-in layers), `postProcessing` (whether this camera runs the post-process chain), `debugChannel`, plus the behavior slot below.
Methods: `:lookAt(target)` (entity id string OR `{x, y, z}` table), `:render()`, `:capture()`, `:setTargetTexture(tex?)` (a `renderer.texture.create` handle to render into, or nil for the viewport).
```luau
entity(id).component.add("Camera", { fov = 90, priority = 10 })
entity(id).component.get("Camera"):capture()
```
The scene's play-mode camera is reached as `layers.active.camera`, a handle that
reads and writes the fields above on whichever entity currently carries them.
`layers.active.camera.entity` is the entity ref for that camera and
`layers.active.camera.entityId` its id string, so a script that needs to attach
something to the camera — an `AudioListener`, a child entity — goes through the
ref:
```luau
local cam = layers.active.camera
cam.fov = 70 -- the camera's settings
cam.entity.component.add("AudioListener") -- the entity carrying them
```
## How the camera moves: `behavior` and `follow`
A `Camera` does not move itself. `behavior` names a component that does, and setting it attaches that component to this entity. Clearing it detaches whatever was attached.
```luau
local cam = entity(id).component.get("Camera")
cam.behavior = asset.ref("@builtin::controller.orbital_follow", "component")
cam.follow = playerBody
```
The attached behavior is an ordinary component on the camera's entity, so it is tuned the way any component is: read it with `component.get` and set its fields there. A `component.add` of the same type configures the instance the slot attached instead of adding a second one: the values it passes are written onto that instance and its proxy is returned. That add is the form a scene `build.luau` tunes the behavior with, because its values are saved with the scene and the loaded Camera's behavior takes them back. A key the component reads only as it is created (one of its `initKeys`, a private field) refuses the add and names the key.
```luau
cam.behavior = asset.ref("@builtin::controller.third_person_follow", "component")
local rig = entity(id).component.get("third_person_follow")
rig.cameraDistance = 6
-- the same, as an add of the type the slot attached
entity(id).component.add("third_person_follow", { cameraDistance = 6 })
```
`follow` is the standard slot every shipped behavior reads. Set the follow target on the **Camera**, not on the behavior, so swapping behaviors keeps it. A behavior that finds its own `follow` field empty falls back to this one, which is what lets a rig keep tracking the player across a behavior swap.
`followResolves` answers whether that slot names an entity that is live — `true` while it names a live one or names nothing at all, `false` once the target is despawned or the id names no entity. A rig whose target does not resolve holds its last pose, and this is the field that tells it from a rig posed correctly on a subject that has not moved. `camera.get` carries the same value beside `follow`, and a write naming an id with no entity behind it draws a warning where it lands.
```luau
cam.followResolves -- false once the followed entity is gone
tools.use("camera", "get", id).followResolves -- the same answer off the tool
```
The shipped behaviors live under `@builtin::controller.*`: `orbital_follow`, `third_person_follow`, `first_person`, `free`, `orbit`, `chase`, `isometric`, `rts`, `birds_eye`, `side_scroller`, `cinematic`, `menu`.
## Writing your own
Any component can be a camera behavior. Write one that moves its own entity and attach it the same way:
```luau
cam.behavior = asset.ref("MyChaseCam", "component")
```
`follow` lives on the Camera, so a behavior gets no property notification of its own when the target changes. Declare `onFollowChanged(newFollow, oldFollow)` to be told the moment it does — the Camera calls it on the component it attached, which is what lets a rig re-pose on the new subject at once. A behavior that reads `Camera.follow` on its own schedule declares nothing and is attached the same way.
```luau
typed function public:onFollowChanged(newFollow: any, oldFollow: any)
-- pose this entity against the new target
end
```
To make it appear in the discovery catalog alongside the shipped ones, declare the `cameraBehavior` tag in the component's `.metadata`:
```json
{ "tags": ["cameraBehavior"] }
```
The tag governs **discovery**, not attachment. A tagged component is listed by `layers.active.camera.behaviors` and is what tooling offers when something asks "which camera behaviors exist"; an untagged component attaches just as well and simply stays out of that list. Tag the ones you want other people (and agents) to find.
```luau
for name, ref in pairs(layers.active.camera.behaviors) do
print(name, ref.identity)
end
```
## Texture colour space
`textureColorSpace = "display"` applies the display transform when rendering into a texture. `"linear"` writes scene-linear values instead. `postProcessing` independently controls the effects chain in either mode. Floating-point targets retain values above one; normalized targets clamp to their representable range. The main viewport uses display encoding.
# Lit Checkerboard
A static, lit checkerboard surface built on `@builtin::shaders.lit_checkerboard` — the default ground material for newly seeded worlds. It uses a neutral light-grey base colour (`0.85, 0.85, 0.85`) with a grey emissive tint (`0.35, 0.35, 0.35`) so the pattern stays readable in dim scenes, high roughness (`0.7`), no metallic, opaque, and instanced.
## Inputs
- `colors.base_color` (`0.85, 0.85, 0.85, 1.0`) — neutral grey checker palette.
- `colors.dark_color` (`0.35, 0.35, 0.35, 1.0`) — the dark square colour, close enough to the base that the pattern stays readable in low light.
- `floats.roughness` (`0.7`) — mostly diffuse surface.
- `render.instancing` (`true`) — supports instanced draws.
# Asset
Promote the bundle's spawned children from temporary to permanent scene state and remove the Asset component, "baking" the bundle's contents directly into the owning scene. Subsequent saves persist the baked entities verbatim and stop replaying the bundle template.
Location: `src/lua/lib/components/Asset.component`
# PlayerSpawn
A world entity that spawns players from a prototype. It carries the fields describing how spawning happens: which `prototype` to instantiate, the `team` and `role` assigned to spawned players, a `maxPlayers` cap (0 = unbounded), the `spawnPolicy` (how many players spawn from this point), and the `placement` (where a spawned player is positioned). The entity's own transform is the spawn point when `placement` is `at_spawn_transform`. In edit mode it draws an orange silhouette outline at the spawn point and surfaces its own setup-validation messages (missing or mis-targeted prototype) on the component's error channel.
Location: `src/lua/lib/components/PlayerSpawn.component`
# lamp_post (bundle)
A lamp post: a slim post, a lit head sitting on top of it, and the point light
it casts. Four entities under one root — `post`, `head` and `glow` — built from
builtin meshes and materials, so the bundle carries no geometry of its own.
Its foot sits at `y = 0` and it stands 2.53 units tall, so placing it at a
position on the ground puts it on the ground.
```lua
asset.resolve("@builtin::bundles.lamp_post", "bundle"):instantiate(nil, {
position = { 4, 0, 2 },
rotation = { 0, 30, 0 },
scale = 1.2,
})
```
## What it is for
It is the prop a starting scene places to show what placing an asset does: one
call brings a whole hierarchy in as the scene's own entities, and the `Light`
riding along shows that a bundle carries behaviour and not only shape.
Reach for it whenever a scene needs a light source that reads as an object
rather than as a floating value.
# orbital_follow
Three-ring orbit camera. Framing is authored as three rings around the follow target — a bottom, a centre and a top, each with a `height` above the target's origin and a `radius` out from its axis. Vertical look slides the camera along the spline through those rings; horizontal look orbits around them. The shape of the orbit is geometry you can see and edit rather than a pair of pitch limits you cannot.
Public fields: `follow`, `ringBottom` / `ringCenter` / `ringTop` (each `{ height, radius }`), `verticalAxis` (0 at the bottom ring, 1 at the top), `horizontalAxis` (the world heading the camera looks along, in radians about +Y from world +Z), `radiusScale`, `aimHeight` (the aim point above the target's origin), `damping` (per-axis time constants in seconds of the camera's offset from its subject: they soften a change of framing, a look, a zoom or a pull-in, and the subject's own motion carries the camera at any speed), `cameraRadius` (radius of the pull-in sweep; `0` turns the pull-in off), `cameraIgnore` (entities whose colliders the pull-in passes through), `sensitivity`, `invertY` (off by default: pushing the mouse forward looks UP, drawing the camera toward the bottom ring; on pitches the other way), `zoomSpeed`, `minRadiusScale`, `maxRadiusScale`.
The rig owns that heading. Binding a follow target aims it along the subject's facing, so the shot opens from behind whoever the camera is told to follow, and look input turns it from there — the subject's facing is never read again. That is what makes the camera steerable for a character who turns to face where it is walking: hold "forward" and the character walks the line the camera looks down, instead of chasing a heading that moves as it turns.
The default rings frame a roughly 1.8 m humanoid standing at its origin: the centre ring sits above head height at the widest radius for a resting shot from behind and slightly above, the bottom ring drops to knee height and pulls in to look up, and the top ring climbs and tightens for a near-overhead look down.
Writing any framing field re-poses the rig immediately with damping bypassed, so editing a ring moves the camera to the shot that ring describes.
`follow` falls back to the `Camera.follow` slot when it is empty, so a rig selected through `Camera.behavior` tracks whatever that camera already follows.
Controls: mouse to orbit and to slide along the rings (play mode requests pointer lock, so no button is held), scroll to scale the ring radii.
```luau
local cam = entity(id).component.get("Camera")
cam.behavior = asset.ref("@builtin::controller.orbital_follow", "component")
entity(id).component.get("controller.orbital_follow").ringCenter = { height = 2.6, radius = 6 }
```
## The camera pulls in when a collider is in the way
Each frame the rig sweeps a sphere of `cameraRadius` from the point it aims at out to where its framing puts the camera, and stands the camera where the sphere first meets a collider. Every collider in the world stops that sweep, whether or not anything draws it: an invisible walk bound (a `BoxCollider` with no `Model`), a trigger volume and a solid wall all pull the camera in alike. The follow subject is the one entity the sweep passes by default.
So a camera that snaps in close to the player, or collapses onto it, while nothing visible stands behind the player is meeting a collider nobody can see. Two fields settle it:
- `cameraIgnore` lists the entities whose colliders the camera sees through. Name the walk bounds and gameplay volumes there and the sweep passes them as it passes the subject, while walls the player can see still pull the camera in. Each entry is an entity (or its id), and it is the collider's own entity that is listed.
- `cameraRadius = 0` turns the pull-in off: the camera stands where its framing puts it, through every collider.
```luau
local rig = entity(camId).component.get("controller.orbital_follow")
rig.cameraIgnore = { entity.find("NorthBound"), entity.find("SouthBound") }
-- or drop the pull-in entirely
rig.cameraRadius = 0
```
## Attach through `Camera.behavior`
The Camera owns the behavior slot, so assigning the rig there is what lets the Camera tell it when `follow` changes. A rig added straight onto the entity with `component.add` still poses, but the Camera does not know about it and a `follow` assigned afterwards will not reach it.
## Player cameras: leave `follow` unset
Entering play drops and respawns `PlayerPrototype` subtrees, so the body is recreated with a NEW entity id. A `follow` authored on the rig keeps pointing at the original, which is no longer live, and the camera freezes at its last pose instead of tracking the player.
For a player camera, set the target on the **Camera** and leave the rig's own `follow` empty:
```luau
entity(rigId).component.get("Camera").follow = body -- PlayerPrototype rebinds this on every spawn
```
`PlayerPrototype` refreshes `Camera.follow` to the live body each time it clones, and the rig reads that slot whenever its own `follow` is empty. Set the rig's `follow` only for a camera that watches a fixed, non-respawning subject.
# Plastic Red
A glossy red plastic — a non-metallic PBR surface with a saturated red base colour (`0.8, 0.1, 0.1`) and mid roughness (`0.5`) for a soft dielectric sheen. Built on `@builtin::shaders.pbr`. Opaque, back-face culled, queue `2000`.
## Inputs
- `colors.base_color` (`0.8, 0.1, 0.1`) — the red plastic tint.
- `floats.roughness` (`0.5`) — soft, semi-glossy highlight.
- `floats.metallic` (`0.0`) — dielectric (non-metal) response.
- `floats.emissive` (`0.0`) — no self-illumination.
# UI Templates — main scene
The startup scene for the **UI Templates** world. `scene.json` materializes a
ground plane, ambient + directional lighting with a procedural sky, and the
standard multiplayer player setup (a `PlayerPrototype` with body + `CameraRig`
under `PlayerSetups`, and a `PlayerSpawn` under `Spawns`). `entrypoint.luau`
holds the lifecycle callbacks (all commented out by default).
This world collects reusable, ready-to-read **UI screen templates** authored
against the `ui.*` CSS-parity surface. Each template is a self-contained script
under `/source/ui/` that registers and shows an `ui` screen. Current templates:
- `ui/p5_menu.luau` — a Persona 5 main-menu lockup: a font-family list resolving
the bundled **Arial Black** face for the blocky title, clip-path polygons for
the torn wedge / octagon badges / row tears, CSS transforms, `-webkit-text-stroke`,
text-shadow, and a repeating-linear-gradient halftone field. Run it with
`execute { file = "/source/ui/p5_menu.luau" }` to display the screen.
# Tessellation Showcase
Every surface in this scene starts as a **flat two-triangle plane**. The
`Tessellation` component subdivides each triangle into thousands of GPU
micro-triangles and displaces them along their normals from a **height map**,
turning flat geometry into rich relief — no authored high-poly meshes.
The entrypoint drives the same component with four procedural height maps:
| Piece | Height source | What it shows |
| --- | --- | --- |
| Rocky mountains | 5-octave value-noise fBm | Natural terrain from a flat plane |
| Sand dunes | Low-frequency swells | Smooth rolling forms |
| Ocean swell | Crossed sine waves | Repeating wave crests |
| Before / after | Bold ridge grid | Same box, flat vs. tessellated |
## What it demonstrates
`Tessellation` sits next to a `Model` + `Material`. It reads the entity's
current mesh, subdivides each triangle into `4^level` micro-triangles, and
displaces them from the material's displacement map — the entity's PBR
material renders the result. A single flat box + a height map becomes a
detailed surface entirely on the GPU.
## Running
Load the scene and enter play mode. The height maps are generated
procedurally on load, so the showcase is fully self-contained.
# assetType runtime
The machinery that reads a `<typename>.assetType/behavior.luau` and turns its
declarations into live behaviour.
A `behavior.luau` is a table of declarations — `ref` methods, `modules` shared
code, an `events` schema, a `namePattern`, `refShapes`, and the `onCreate` /
`onRegister` / `onChange` / `onDelete` / `validate` / `settingsDefaults` /
`settingsCorrections` hooks. None of it runs itself. This is the code that
runs it, and it lives here because reading a type's declarations is the
**assetType type's own behaviour** — the same way `material.assetType` owns
what a `.material` does.
## Sub-modules
| Module | Reads | Responsibility |
|---|---|---|
| `ref.module` | `ref`, `modules` | Builds every `AssetRef` envelope, attaches the shared metatable, and dispatches per-type methods and shared modules onto it. Also owns edit-mode write-behind persistence. |
| `create.module` | `onCreate`, `namePattern`, `validate` | Backs `asset.create`: resolves the type, checks the name against its pattern, runs `onCreate` (or clones `template/`), and writes the instance. `validate.module` beside it checks a creation `opts` table against the type's `createSchema`. Before running `onCreate`, resolves the type's settings chain (below) and hands the result to the hook as `opts.settings`. |
| `settings.module` | `settings`, `settingsDefaults`, `settingsCorrections` | A type's declared settings schema and where its values live (`type.yaml`'s `settings:` block). `Settings.resolveForCreate` layers the schema's own defaults, the type's own `settingsDefaults(name, opts, asked)` hook (a DEFAULT the schema can't express — a name-derived convention, a fact read off a payload the caller handed it, a value following a field `asked` shows a higher layer naming; any later layer overrides it), the Preset Manager, an importer's settings, the caller's `settings` table, then the type's own `settingsCorrections(values, name, opts)` hook (a CROSS-FIELD CONSTRAINT the schema can't express, applied last over the fully merged values). A correction over a default/preset/importer value lands silently; a correction that would override what the caller's own `settings` table explicitly asked for refuses the creation instead, naming the field and why. Neither hook is ever handed a table to mutate; each returns what it wants changed (`settingsCorrections` may also return a `{ [field]: reason }` table for that refusal message). `Settings.cloneValues` deep-copies a values table — `create.module` hands `onCreate` a copy, never the table it stamps. `Settings.valuesOf` / `writeValues` read and write an asset's own values file. |
| `typeYaml.module` | `type.yaml` | The folder a type lives in and its parsed `type.yaml`, decoded once per content version of the file. `settings.module` and `inspector.module` read their blocks through it. `about` / `aboutAt` read a type's `about.yaml` (`intent`, `function`, `usage`), leaving out any key that is empty or still starts with `[[describe]]`; `aboutRows` and `aboutMarkdown` lay the texts out for the Inspector and the `types/<name>` guide. |
| `inspector.module` | `type.yaml`'s `inspector:` block | Which inspector a type declares for its assets (`inspector: { type: ... }` or `{ instance: ... }`), and the declaration that file returns — the sections it adds, the generic ones it replaces, and their order — or the problem that kept it from loading. |
| `onRegister.module` | `onRegister` | Fires an instance's once-only registration hook — on arrival, on world load, and for `@builtin` library instances. |
| `changeDispatch.module` | `onChange`, `onDelete`, `refShapes` | Routes a VFS source write or a folder delete to the enclosing typed asset's type hooks. |
| `refShapes.module` | `refShapes`, `ref` | Publishes what every `AssetRef<category>` answers to, so a member read on an asset-typed value is checked against that category's real surface. |
| `events.module` | `events` | The per-asset event runtime behind `ref.events`. |
| `instantiable.module` | `ref.instantiate` | The instantiation contract an asset type opts into: `root` stands the root entity, `place` applies the base placement opts to one the type adopted, and `result` shapes the `(root, idMap)` every `instantiate` returns. Also registers the `sceneInstantiable` field-constraint validator. |
| `declaration.module` | `type.yaml` | The type's decoded declaration, read once and held until a write inside an `.assetType` folder; what every question about a type as a whole reads. |
| `previewRecipe.module` | `type.yaml` `preview:` | How an instance's `preview.png` is produced: which of the four recipes (`instantiate`, `image`, `surface`, `custom`) the type declares, and what that recipe needs. |
## Reaching it
```luau
local rt = require("@builtin::assetTypes.assetType.shared")
rt.ref.loadTypeBehavior("material")
```
Fields resolve lazily. The prelude requires the sub-modules directly, in
dependency order (`ref`, then `changeDispatch`, `onRegister.install()`,
`refShapes.install()`) — a table that required all six on load would take that
ordering away from it.
# io.engine (procNode)
Call a function of the engine bridge on the CPU: reading an asset as a graph value, writing a value as an asset, reading scene entities as columns, and reading the scene's water.
Built from the Luau node (`io.luau`) over `engine.module`, which sits in this node's folder. `core.entity.read`, `entity.transform`, `scene.water`, `sim.scene_colliders` and `io.export` call it with the function they need; `io.import` calls `engine.module` through `io.luau`. Every function answers plain Luau tables: a column of values is a list of numbers, a list of `{ x, y, z[, w] }` records, or `{ type = "String", values = { ... } }`, which the caller hands to `op.transfer.upload`.
## Inputs
- `fn` (String, default "importAsset", one of "importAsset", "writeAsset", "countEntities", "readEntities", "importWater"). The bridge function to call. `importAsset` takes an AssetRef, as `asset.resolve("<identity>", "<type>")` writes one, and the options its type reads it with.
- `args` (List<Value>, optional). The values the function is called with, in order.
- `returns` (List<TypeName>, optional). The procgen types of the values the function returns, in order.
- `ceilingMs` (Float, default 0, at least 0). The CPU time the function may run for, in milliseconds, counting the work it runs in any coroutine it resumes and in any task it starts while the call lasts; a call past it is refused. The time it waits on a frame, a program build, a read-back or a task, and the time its thread waits for a core on a busy machine, are not counted. 0 leaves the runtime's default ceiling.
- `live` (String, default ""). The name of a bridge function that answers a digest of the state `fn` reads, such as `readEntitiesLive` for `readEntities`: the graph's live digest folds it in, so a Generator cooks the graph again when the scene state moves. Left empty, the call reads nothing the graph does not hold.
## Outputs
- `value`: what the called bridge function returned.
```luau
local read = g.io.engine({ fn = "importAsset", args = { asset.resolve("@builtin::meshes.cube", "mesh"), {} }, returns = { "Value", "Record" } })
```
# scene_saver.recordDiff
The comparisons the saver's record diffs (`transformDiffLines`, `recordDiffLines`, the pending scene edits) are built from.
- `diffEqual(a, b)` and `valuesEqual(a, b)` compare two record values, numbers within a tolerance that absorbs the f32 round trip of a live read.
- `componentMatchName(t)` answers the name a component type matches by, whichever spelling the record used; `componentList(rec)`, `componentMatchNameOf(c)`, `componentInstanceLabel(c, typeName)` and `componentData(c)` read a record's components the way the comparison needs them, answering empty for a malformed record.
- `selfWrittenFieldSet(entityId, componentType, instanceName)` answers the fields a component instance wrote about its own runtime.
- `pushAttributeDiffLines(oldRec, newRec, push)` hands `push` one line per attribute added, changed or removed.
```luau
local RecordDiff = require("@builtin::assetTypes.scene.shared.saver.recordDiff")
assert(RecordDiff.diffEqual({ 1, 2 }, { 1.00001, 2 }))
RecordDiff.pushAttributeDiffLines({ attributes = { hp = 1 } }, { attributes = {} }, print) -- - attribute.hp
```
# scene_saver.serialize
One live entity as the record a scene file carries, and the reads the saver asks of a live entity while it decides what to record.
- `serializeEntity(id, parentById)` answers the record for a live entity: its name, parent reference, transform, components (each field value serialized), participation, network scope, render layer and attributes. A field holding a runtime GPU handle is reported once, with the remedy, and left out of the record.
- `serializeFieldValue(v)` answers one component field value as a record states it, with the reason when the value has no record form.
- `entityDisplayName(id)` answers the name an entity is shown by, and `entityIsTemporary(id)` whether it is marked as never saved.
```luau
local Serialize = require("@builtin::assetTypes.scene.shared.saver.serialize")
for _, e in ipairs(entity.findAll()) do
print(Serialize.serializeEntity(e.id).name)
end
```
# Component (asset type)
A component is a reusable behavior + data unit attached to an entity.
The component file declares public fields, methods, and lifecycle
callbacks; the engine instantiates one per attachment and drives its
hooks every frame. Components are the primary unit of gameplay code —
movement, AI, animations, UI panels, network sync, and per-entity state
all live here.
## When to use one
- You need per-entity state that survives across frames.
- You need per-entity behavior driven by engine lifecycle hooks
(`start`, `update`, `onDestroy`).
- You want to expose configurable fields to designers / `entity.spawn`
callers / scenes without code changes.
If you need a stateless pure-function library, use a `.module`. If you
need to generate content declaratively, use a `.service`; compute work
belongs in a `.computeShader`.
## Where it lives
- Source: `/zero/source/.../<Name>.component/`
- Identity: `<Name>` (the `.component` suffix strips from the
identity, but the folder retains the suffix on disk).
- Folder shape:
- `init.luau` (or `init.lua`) — declares the component. **Required.**
- `README.md` — instance-level documentation. **Required.**
- `.metadata` — agent-editable tags + free-form fields. **Required.**
The structural contract is enforced by `asset.validate` against this
type's `type.yaml`.
## How to create one
Always start with the scaffold — it guarantees the folder shape
matches `type.yaml` from the moment it's on disk:
```luau
asset.create("component", "<Name>")
-- Creates: /zero/source/<Name>.component/
-- init.luau (canonical declare + public block + hooks stubbed)
-- README.md (instance README template)
```
After scaffolding, edit `init.luau` to fill in the actual fields and
behaviour. The component registers as soon as the file is written —
hot-reload is automatic; no engine restart required.
## How it operates
1. **Registration.** Writing `init.luau` into a `.component/` folder
indexes the asset, mints a guid (in the `.meta` sidecar), and
surfaces the type to the engine's component registry. Same-frame
visible to `entity.component.add`.
2. **Attachment.** `entity(id).component.add(name, data)` (or the
batch form `tools.use("entityOps", "addComponents", targets,
{ [name] = data })`) attaches an instance.
The `data` table's fields override the component's declared
defaults; omitted fields take the declared default. The `data`
values are copied into `public.*` BEFORE `awake()` runs.
3. **Lifecycle.** The hooks run in this order:
- `awake()` — same frame as attach, immediately after the data
copy. Initialise state, set up native bridges, register with
other systems. Other components on the same entity may not have
awoken yet.
- `start()` — the next frame. Every component on the entity has
completed `awake()` by then, so cross-component setup is safe
here.
- `update(dt)` — same frame as `start()`, then every frame after,
for as long as the gameplay clock is running. `fixedUpdate(dt)`
runs once per physics step, straight after it, with the step's
`dt`, so a frame can run it several times or not at all;
`editorUpdate(dt)` runs
while the clock is held. Which of them ticks is decided by
`engine.mode` AND `engine.paused` together — see **Per-frame
hooks and the two axes** below, and `man modes`.
- `onDestroy()` — when the component is removed via
`component.remove` or when the owning entity is despawned.
Release resources, drop native ECS components.
For the full list (`onEnable`, `onDisable`, `fixedUpdate`,
`editorUpdate`, `onPropertyChanged`, `onAssetReload`,
`onModuleReload`,
`onBeforeSave`, …) and the timing details, **read `man components`
— that is the canonical lifecycle reference. Generic component
patterns (declare blocks, public fields, sync, ECS bindings,
typed methods) all live there.**
4. **Public surface.** Fields declared in `public` are inspectable,
editable in the inspector, and persist with the entity. Methods are
declared on that same table with a colon —
`typed function public:method(args)` — and every one of them takes
the component ref as `self`, so a caller invokes it with a colon
too: `entity(id).component.get(name):method(args)`. That is the one
calling convention a component ref answers to: the reflection
methods `ref:fields()` / `ref:getField(name)` /
`ref:setField(name, value)` take the same spelling. A function the
chunk declares at file scope (`function helper()`) is the
component's own, reachable from its hooks under its bare name;
`public` is what the ref carries.
5. **Settings and the Inspector card.** `type.yaml` names the component's
settings schema as `settings: { schema: { instance: "init.luau",
export: "public" } }`: the `public` table IS the schema, and its values
live on each entity the component is on, so `asset.schema(componentRef)`
answers the Field descriptors and no values file exists. On an entity
the Inspector shows the component as a card whose generic `settings`
section edits those fields as typed controls, each edit checked against
the schema and written to the live component as one undoable operation.
`type.yaml` also names `inspector: { instance: "inspector.luau" }`: a
component that ships an `inspector.luau` beside its `init.luau` shapes
its card — the file returns `{ sections, replace, order }` (see
`assetTypes/README.md`, "`inspector:`"), each section's `build(ctx)`
reading `ctx.component` (the live component) and `ctx.selection.id` (the
entity). `SceneModule.component/inspector.luau` replaces `settings` with
the bound module's declared inputs.
6. **Events.** An `events` block declares typed moments the component
fires from inside — `events = { onHit = Event({ dmg =
Field.number(0, NoSync) }) }`. The owner fires with
`events.onHit:fire({ dmg = 10 })`; outside code subscribes through
the proxy's subscribe-only facade,
`entity(id).component.get(name).events.onHit:connect(fn)`
(`:once` / `:wait` too). Pass `Sync` as the event's second argument
to replicate a fire to other peers. Events let one component react
to another's moments without polling. Full reference: **`man
components`**, Events section.
7. **Hot reload.** Editing `init.luau` reloads the component
definition; live instances are reconciled with the new declaration
in-place, and each crosses one lifecycle boundary: `onDestroy()`
ends the running life, then `awake()` and `start()` construct the
one that replaces it, and `update()` resumes on the next frame.
`public` and `private` values carry across; module-level `local`s
belong to the chunk, so the new life starts with them at their
initial values and its `awake`/`start` are what fill them in.
Whatever the old life handed elsewhere — a subscription, a shared
registration, a spawned entity — is released in `onDestroy`, the
call that pairs with the construction about to run. An edit that
leaves the source byte-identical crosses no boundary at all.
A field carries its value across when something stated it: a key of
the `data` table the instance was attached with (a scene record's
data is that table), or a write made since. Every other field takes
the edited source's declared default, so changing a default reaches
every instance that left the field to it. An instance that states
its own value for a field whose default the edit changed keeps that
value, and the reload logs a `[HotReload]` warning naming the field,
the entity, the value kept and both defaults.
## Discovery
- `asset.list("component")` — every registered component (engine +
world + libraries).
- `asset.inspect("<name>")` — identity, scope, public fields,
lifecycle methods, source path, and this type README. `detail`
carries the component's declared `fields`, `methods`, and `events`,
parsed from its source — `tools.use("assets","describe", "<name>")`
renders it as markdown, so a component's events are visible without
opening its source.
- `ls /zero/source/<Name>.component` lists the files the folder holds.
- `entity(id).component.list()` — components attached to an entity.
## Per-frame hooks and the two axes
`engine.mode` (`"edit"` / `"play"`) and `engine.paused` (the gameplay
clock) are independent, and the per-frame hooks read them separately:
- `update(dt)` / `fixedUpdate(dt)` tick while **the gameplay clock is
running** — whichever mode the engine is in.
- `editorUpdate(dt)` ticks while **the clock is held** — because the
mode is `edit`, or because play is paused — and also alongside
`update` whenever edit is running.
Which gives four states, and this table is the contract:
| `engine.mode` | `engine.paused` | hooks that tick |
|---|---|---|
| `edit` | `true` | `editorUpdate` |
| `edit` | `false` | `update`, `fixedUpdate`, `editorUpdate` |
| `play` | `false` | `update`, `fixedUpdate` |
| `play` | `true` | `editorUpdate` |
Entering edit sets `paused = true` and entering play sets it back to
`false`, so a session that never writes `engine.paused` only ever sees
rows 1 and 3 — which is where the shorthand "`update` is the play tick,
`editorUpdate` is the edit tick" comes from. It is the two default rows
of a four-row table, not the rule.
**Row 2 is the one that surprises.** With `engine.paused = false` in
edit — set by hand, or by a tool that advances the gameplay clock to
sample it, such as a `capture.collage` given a `duration` — `update` and `editorUpdate`
both fire, once each, every frame. A component that declares both and
routes them to one body therefore runs that body **twice per frame**:
```luau
local function drive(dt) ... end
function update(dt) drive(dt) end
function editorUpdate(dt) drive(dt) end -- twice per frame in row 2
```
Nothing reports the doubling; it reads as an animation running at twice
its intended speed. To get exactly one tick per frame in every row, ask
which axis you meant and answer it in the body:
```luau
function update(dt) drive(dt) end
function editorUpdate(dt) if engine.paused then drive(dt) end end
```
A scene `entrypoint.luau` declaring the same two hook names is dispatched
by the scene loader, not by this table: there `update` needs play AND
running, `editorUpdate` needs edit, and play-paused ticks neither. Read
`man scenes` before carrying a component's rule to an entrypoint.
## Authoring conventions
- Annotate public fields with `--!desc` so the inspector + LSP show
documentation alongside the field name.
- Declare lifecycle hooks (`start`, `update`, `onDestroy`) explicitly
rather than relying on engine defaults — readers can see at a glance
what runs when.
- Keep `update(dt)` cheap — it runs every frame per instance. Move
heavy work to `start`, to a `.computeShader`, or behind a job (`man jobs`).
- For multi-entity batching, prefer
`tools.use("entityOps", "addComponents", targets, components)` over
a loop of `entity(id).component.add` — it takes many targets and
many components in one call.
## Common pitfalls
- **Naming.** The component identity is the folder stem with
`.component` stripped (e.g. `<Spinner>.component/` → identity
`Spinner`). Don't repeat the suffix in the identity passed to
`component.add`.
- **`init.lua` vs `init.luau`.** Either is accepted (it's a
`one_of_group: "entrypoint"`), but `.luau` is canonical. Mixing both
in one folder is invalid.
- **`update` in edit mode.** It runs whenever the gameplay clock is
running, and entering edit stops that clock — so `update` is idle in
a freshly-entered edit session and ticks again the moment anything
sets `engine.paused = false`. Use `tools.use("wld", "play")` to test
update logic under the mode it ships in; `tools.use("wld", "edit")`
to return. See **Per-frame hooks and the two axes** above.
- **Renaming.** Renaming the folder changes the identity. References
to the component are recorded against the component itself, so
renaming or moving it does not break them.
## Related types
- `.module` — pure-function library, no per-entity state.
- `.service`: a declarative content generator, no per-entity state.
- `.preset` — a captured component configuration, replayable into
`component.add` without re-typing fields.
- `.bundle` — entity hierarchies that may carry component attachments
as part of the bundle.
## In the Inspector
Fields: its execution order, hooks, and how many methods and events it declares, then the fields its `public` table declares as read-only typed controls holding their defaults, with **Check** and **Open** for its entry file. Instances: a link to every entity carrying it. What its card shows on an entity comes from the component's own `inspector.luau`.
# Material (asset type)
A material is a named, configured instance of a `.shader`. The shader
defines the surface look (lighting model, available property fields,
texture slots); the material picks values for those fields and a
shader to bind to. Renderable entities reference materials, not
shaders, so the same geometry can wear different materials and the
same material can be reused across many entities.
Entities sharing one material can still differ in a value: every drawn
object carries four `vec4` lanes of its own that a surface shader reads
as `input.shader_data[lane]` and `renderer.instanceData.set` writes. Use
that for a dissolve at its own progress per subject, a hit flash at its
own age, a fill level per instance — a material per entity is for when
the LOOK differs, not when a number does.
> New to how a material relates to its shader, its textures, and the live GPU
> material behind it? Read `core/resource-model` first. It also covers the
> difference between an authored `.material` asset (this type) and a runtime GPU
> material made with `renderer.material.create`.
## When to use one
- You want a named surface configuration an entity wears by referencing it
from its `Model` component (`material = "@.../<Name>"`).
- You want runtime property tweaks (`asset.resolve(ref):setProperty(...)`) to
affect every entity using the material at once.
- You want to publish a curated look (`Gold`, `RedGlow`,
`BrickWall`) other authors can reuse without re-typing shader
bindings.
If you need a brand-new lighting model or a custom visual effect, you
need a `.shader` first; the material wraps it.
## Where it lives
- Source: `/zero/source/.../<Name>.material/`
- Identity: `<Name>` (the `.material` suffix strips).
- Folder shape:
- `mat.yaml` — canonical material spec (shader + property values).
**Required.**
- `README.md` / `.metadata` — optional prose + tags, indexed when an
author writes real ones.
- `textures/` — optional folder for embedded textures. Validator
allows the folder; its interior is unrestricted.
- `shaders/` — optional folder for a co-located custom shader.
Same rule.
## How to create one
```luau
asset.create("material", "<Name>")
-- Creates: /zero/source/<Name>.material/
-- mat.yaml (shader binding + property scaffold)
-- `folder` places it in a subfolder of /source instead of the root:
asset.create("material", "<Name>", { folder = "materials" })
```
Pass the shader and initial values through `asset.create`, then tune with
`setProperty` and keep what you tuned with `saveDefinition`:
```luau
local mat = asset.create("material", "<Name>", { shader = "@builtin::shaders.pbr" })
mat:setProperty("base_color", { 0.8, 0.6, 0.1, 1 })
mat:setProperty("roughness", 0.3)
mat:saveDefinition() -- writes the tuned values into mat.yaml
```
This mints the `.material` asset and returns a handle you tune with
`setProperty`. A set is a runtime change: it reaches the GPU material and every
entity wearing it, and `saveDefinition` is the call that writes it into
`mat.yaml`, so a property driven every frame costs no file write and leaves the
authored file as its author wrote it. A purely runtime `asset.create` lives only
for the session.
For an ephemeral, runtime-only GPU material that is not an asset at all, use
`renderer.material.create` instead (see `core/resource-model`).
## How it operates
1. **Property reflection.** The bound shader's uniform struct fields
become the material's configurable properties — no manual property
declarations needed.
2. **Property values.** `mat.yaml` groups values by type, each block
keyed by the shader's field names: `properties:` holds scalars and
colors/vectors, `ints:` holds integer-typed fields, `bools:` holds
boolean-typed fields, and `textures:` holds texture slots. Omitted
fields take the shader's default. A scalar is a bare number
(`roughness: 0.5`); a color or vector is an array
(`base_color: [1, 1, 1, 1]`). `shader:` is the shader's identity
string. This is the shape every material the engine writes uses. A
`saveDefinition` writes the changed values into the file where they
already stand and leaves the rest of it — its comments, the order of
its keys and blocks, the spelling of its literals, untouched.
3. **Render state.** `render:` holds how the mesh is drawn rather than
what the shader reads: `blend` (the equation the fragment composites
with, also spelled `type`), `cull`, `depth_write`, `depth_compare`,
`topology` and `queue` (the draw order, a whole number: `2000` opaque,
`3000` transparent, `4000` overlay). The `.shader` type README's "The
render block" section lists every value each key takes, and
`asset.create`'s `render` argument takes the same keys and values.
4. **Application.** An entity wears a material by referencing it from its
`Model` component (`material = "@.../<Name>"`). Materials affect every entity
that references them; runtime `asset.resolve(ref):setProperty(...)` changes
propagate to all of them immediately, and last until the next mode change
unless `saveDefinition` writes them into `mat.yaml`.
5. **Texture binding.** Texture slots accept `color:r,g,b,a` for
solid colors, `default:white` / `default:black` / `default:normal` for the
engine's own fallback texels, `@builtin::textures.<name>` for builtins,
`path/to/image.png` for file textures, or a render-target handle
for camera renders. The GPU texture cache reads the `color:` and
`default:` forms from the string itself, so a slot written in one names no
asset and carries nothing with the material when it travels; every other
form names a texture asset the slot depends on.
A texture in `base_color_texture` is multiplied by `base_color`: every
built-in surface shader declaring both draws the texture's colour times
`base_color`, channel by channel. Under the default white `base_color` the
texture shows as itself, and a coloured `base_color` tints it. The two
read a colour differently: `base_color` is linear, and the slot reads its
texels as sRGB-encoded, a `color:` value's included, the way it reads an
image. A colour stated in both places is therefore multiplied by a darker
copy of itself: `color:0.62,0.14,0.05,1` in the slot decodes to
`0.34, 0.017, 0.0039`, so beside `base_color` `0.62, 0.14, 0.05` the
surface draws `0.21, 0.0024, 0.0002`, far darker than either. State a
solid colour once, as `base_color`, and leave the slot at its default
white.
6. **Hot reload.** Editing `mat.yaml` repacks the GPU uniform buffer
on the next frame; visible immediately.
**Properties that follow a named value.** `values:` binds a property to a
named value of the entity wearing the material (`values.set(entity, name,
v)`, the `topics/values` guide): `property: valueName`, or
`property: { value: "name", scale: s, offset: o }`, and the property draws
`value * scale + offset` on each entity holding the value, read on the draw.
A boolean reads as 1 or 0, and a number fills every channel of a colour or
vector property. The wearer's own value is read, else its nearest
ancestor's, so a lamp mesh under a car lights from the car's `brake`; a
wearer holding none draws the material's own value. A scalar, vector,
colour or bool property binds (at most 32 per material); a texture slot
does not. `asset.create("material", name, { values = { ... } })` and
`renderer.material.create({ ..., values = { ... } }, key)` take the same
map.
```yaml
values:
emissive_intensity: { value: "brake", scale: 3 }
```
7. **Inspector.** `type.yaml` names `inspector: { type: "inspector.luau" }`,
and `inspector.luau` declares the material's sections beside the generic
ones: `shader` (the bound shader, swappable), `properties` (each shader
field as its typed control) and `textures` (each texture slot as an asset
chip). They read the material through `mat:getShaderRef()` (the bound
shader's ref, `~` expanded) and `mat:getTextures()` (slot name →
`{ ref, name }`: the bound texture's guid and the id the material names
it by). An edit there is an authoring write: `mat:writeProperty(name,
value)` and `mat:writeTexture(slot, texture)` write the value into
`mat.yaml` (and drop the session's own value for it, so the material
reads what the file holds), and `setShader` rewrites the file; inside an
undo operation each file write is recorded, so undo puts `mat.yaml` back.
`setProperty` / `setTexture` stay runtime writes that last until the next
mode change.
## Discovery
- `asset.list("material")` — every registered material.
- `asset.inspect("<name>")` — shader binding, property values,
source path, this type README.
- `ls /zero/source/<Name>.material` lists the files the folder holds.
- `asset.resolve("<name>"):getProperties()` — the property fields exposed by
the bound shader, each at the value it currently holds: what `mat.yaml`
authored until something writes to the material, and the written value from
then on, whichever call wrote it. That is the value the surface is drawn
with. Read them back rather than guessing names; which fields exist comes
from the shader.
- `asset.resolve("<name>"):getDefinition()` — the authored `mat.yaml` bytes,
the value the material rests at.
- `asset.resolve("<name>"):getShader()` — the shader the material binds.
## Authoring conventions
- Use lowercase, descriptive names (`brushed_steel`, `red_glow`,
`tarmac_wet`). Avoid shader-name suffixes in the material name —
the binding lives in `mat.yaml::shader`.
- Drive every property the shader exposes; defaults are a fallback,
not a contract for which fields exist.
- Embed textures alongside the material (`textures/` subfolder) when
they're used only by this material; reference shared textures from
`@builtin::textures.*` or a library.
## Common pitfalls
- **Property names must match the shader's uniform struct exactly**
(case-sensitive). Misnamed properties silently get the shader's
default.
- **`base_color_texture` multiplies `base_color`.** A `color:` texture holding
the same value as `base_color` multiplies the colour by its own
sRGB-decoded value, so the surface renders far darker than the value
written. Write a solid colour as `base_color` alone.
- **Broken shaders break the material.** A material whose shader has a
parse error renders as the magenta checkerboard. Inspect the shader
first if the material looks wrong.
- **`setProperty` targets the material, not the entity.** Changes affect
every entity that references the material.
- **`setProperty` is a runtime change.** It reaches the GPU, not the file:
call `saveDefinition` to write the current values into `mat.yaml`. For a
per-entity override that no other entity sees, give the entity its own copy
with `Model:applySessionMaterial(renderer.material.create(...))`.
- **Renaming the folder changes the identity.** A `Model` component or
other reference to the material is recorded against the material
itself, so renaming or moving it does not break them.
## Related types
- `.shader` — defines what properties exist; required before a
material can bind to it.
- `.preset` — captures a component configuration; conceptually
similar to a material, but for components instead of shaders.
# avatar
A playable character, composed from three **independent** parts:
- **body** (optional): a `.bundle` (a skinned `.mesh` + bone entities, a character
imported from a GLB/FBX) or a plain `.mesh`. The look. An avatar with no body is
a controller or first-person camera with no visible mesh.
- **controller** — the movement system. Default: the standard humanoid kit
(`CharacterController` + `MovementState` + `Humanoid`).
- **animation** — the system that animates the body. Default: the shared
`Locomotion` blend space (reads the controller's velocity, plays retargeted
clips). Or `ClipPlayer`. Or your own component.
Movement and animation are **not** tied together — attaching the controller never
pulls in `Locomotion`, and choosing your own `animation` keeps the standard
controller. Authoring your own locomotion is a one-field swap (`animation = ...`).
Because the standard clip set retargets onto **any** humanoid rig, the default
locomotion drives any humanoid body with no per-body setup and no editing of any
built-in content.
## Make an imported character playable
```luau
-- After importing a rigged humanoid GLB, the importer leaves a `.bundle`
-- (skinned mesh + bones). Survey for it, then wrap it in an avatar:
local body = asset.resolve("my_character", "bundle")
local hero = asset.create("avatar", "my_hero", { body = body })
```
To make `hero` the body every connecting player spawns with, point the scene's
`PlayerPrototype.body` at a body entity whose `Asset` source is `hero` (authored
in the scene tree — see `man scene`). To drive the **local** player with it right
now, instantiate it and bind the entity:
```luau
local root = hero:instantiate() -- spawns the body + movement kit
layers.active.players.localPlayer.avatar = root -- bind the live body
```
Only the avatar bound as the local player's body answers the player's
controls. Every other avatar you instantiate (an NPC, a mannequin, a preview)
stands where it is until something drives it: with the standard controller,
write its `CharacterController` input bus (`inputX` / `inputZ` / `inputJump`)
from a component of your own, the way an AI brain does (`man CharacterController`).
```luau
local npc = hero:instantiate() -- stands still
npc.component.get("CharacterController").inputZ = -2 -- walks while this is written each frame
```
An avatar is **not** humanoid-only. The body can be anything — a humanoid, a
vehicle, an unrigged plane with a custom controller and shader. The defaults
adapt to the body: a **humanoid** body defaults to the shared humanoid Locomotion
(the `synty` clip preset); a **non-humanoid / unrigged** body gets **no
auto-animation** (you supply your own `controller` and `animation`). This is the
non-humanoid detection — an arbitrary animated model is never auto-driven by
humanoid locomotion — and it never blocks creating the avatar.
```luau
-- Non-humanoid avatar (e.g. a custom-controlled body), fully generic:
asset.create("avatar", "ship", { body = shipBundle, controller = { system = "ShipController" }, animation = false })
```
## Options
`asset.create("avatar", name, opts)`:
| opt | meaning |
|-----|---------|
| `body` (optional) | the body `.bundle` or `.mesh` ref; omit for a body-less avatar |
| `controller` | movement: `true`/nil = standard kit, `false` = none, or a component name / `{ system, data }` |
| `animation` | a component name or `{ system, ... }` — **author your own locomotion here**; defaults to Locomotion |
| `locomotion` | clip preset for the default Locomotion (default `"synty"`) |
| `clip` | a single `.animation` ref — selects a one-clip `ClipPlayer` |
| `looping` / `speed` / `sourceRig` | `ClipPlayer` tuning (with `clip`) |
| `expressions` | the blend shape each standard expression preset plays, where the body's shapes are not named after the presets: `{ aa = "Mouth_A", blink = { shape = "Eyes_Closed", overrideMouth = "block" } }` |
| `parameters` / `animator` / `menu` / `parts` | the avatar's parameters, its animator, its expression menu and its own parts seated on its body (below) |
| `type` | the avatar type its body is mapped onto (an `avatarType` ref or identity); the builtin humanoid when omitted |
Author your own locomotion (keeps the standard movement):
```luau
asset.create("avatar", "my_hero", { body = body, animation = { system = "MyLocomotion" } })
```
`avatar.json`:
```json
{
"body": { "__ref": "<bundle guid>" },
"controller": true,
"animation": { "system": "Locomotion", "preset": "synty" }
}
```
The body's facing is read from its rig (the importer records a `forward` for each
body — `{0,0,1}` for a +Z Synty/FBX body, `{0,0,-1}` for a -Z glTF body), so
Locomotion turns the body to its movement direction regardless of how its source
authored it. No per-avatar facing knob.
## Use
```luau
local root = avatarRef:instantiate() -- spawn body + movement kit
avatarRef:setCharacter(root, asset.resolve("knight", "bundle")) -- swap the body, keep the behaviour
```
`:instantiate()` spawns the body and attaches the movement system bound to it.
`:setCharacter(root, body)` despawns the current body and spawns a new one in its
place; the Locomotion kit on the avatar root re-resolves the new body's rig, so
the look swaps while the behaviour is kept. Items seated with
`Humanoid:attach` are kept: each moves to the new body's socket for the same
point, with the transform it was seated with. An item whose point the new body
lacks stays on the avatar root and is seated by the next swap to a body that
has it. The new body takes the old body's id and is composed with the options
the avatar was instantiated with, and the avatar's `Asset` id map names it.
## The avatar standard
A humanoid avatar is built around one standard (the `avatarStandard` module):
the VRM 1.0 bones, 19 attachment points, the viewpoint, the face expressions and
a height band. A game relies on it to seat its own items on any visitor's body
(`Humanoid:attach("rightGrip", sword.id)`) and to read where the eyes are.
```luau
local d = avatarRef:descriptor() -- bones, points, viewpoint, expressions, height
print(d.height, d.points.hat.position.y, d.viewpoint.source)
local v = avatarRef:validate() -- { ok, problems }
for _, p in v.problems do print(p.rule, p.name, p.message) end
```
`:validate()` names every rule the body breaks, one problem per missing item: a
required bone (`rightHand`), an attachment point the body cannot carry
(`rightGrip`), a face expression no blend shape answers (`aa`), a height outside
the band, or a body that is not a humanoid at all. A game holds bodies to the
rules it relies on: `avatarRef:validate({ rules = { "type", "bone", "point", "height" } })`
leaves out the face expressions. A live avatar answers the same through its
`Humanoid` (`:descriptor()`, `:validate(opts)`).
## Parameters, an animator and an expression menu
An avatar declares its own state as data: `parameters` (named bools, ints and floats,
synced or local, saved or not, the synced ones 256 bits in all), an `animator` (layers of
state machines over the parameters, whose states pose blend shapes and bones, show and
hide parts, set shader data and play `.animation` clips on the body under the layer's
mask, a wave on the right arm or a dance over the locomotion), a `menu` (buttons, toggles, sub-menus, radial and
two-axis puppets) the person uses from the app menu's **Expressions** tab, and `parts`
(meshes seated on its body's attachment points on every peer). No script is
involved, and every peer draws the same avatar from the parameters it receives.
```luau
local hero = asset.create("avatar", "my_hero", {
body = body,
parts = { { name = "Hat", mesh = "@builtin::meshes.cone", point = "hat", scale = 0.25 } },
parameters = { { name = "Hat", type = "bool", default = true, saved = true } },
animator = { layers = { {
name = "Hat",
states = { { name = "On" }, { name = "Off", motion = { hide = { "Hat" } } } },
transitions = { { from = "On", to = "Off", when = "!Hat", duration = 0 },
{ from = "Off", to = "On", when = "Hat", duration = 0 } },
} } },
menu = { { type = "toggle", label = "Hat", parameter = "Hat" } },
})
local check = hero:animator() -- { ok, problems, summary, syncedBits, cost }
```
`hide = { "Hat" }` names the part the avatar declares; any entity under the avatar is named
the same way, and an attachment point's name (`hat`) names every item seated there. An avatar whose synced parameters pass 256 bits is refused when it
is created, with the overrun stated. The `avatarAnimator` module's README has the whole data
form, and the `AvatarAnimator` component runs it.
## In the Inspector
Selecting an avatar opens its **Avatar** section, read from `avatar.json`: its label when it has one, how it moves (`controller`: the standard mover, none, or the system it names), what animates it, and whether it is humanoid. It links the body the avatar wears, or says the avatar spawns no visible body, and offers to open `avatar.json` in the Code panel.
## The body's skeleton
`avatarRef:rig()` is the `.rig` of the body's skinned mesh (nil for a body-less avatar or a
plain mesh), and `avatarRef:avatarType()` the avatar type its body is mapped onto. The
`avatar` toolbox reads and corrects that mapping (`avatar.inspect`, `avatar.map`), and makes,
dresses and saves avatars; `guides { path: "core/avatars" }` walks through it.
# Bundle (asset type)
A bundle is a composable, spawnable entity assembly— a saved entity
hierarchy plus its components, sub-assets (meshes, textures,
animations, materials, sub-bundles), and any nested data, packaged
into a single addressable identity. Bundles are the engine's "prefab"
primitive.
There are **two ways** to bring a bundle into a scene — and the
difference matters enormously for scene saving and live editing. See
"How to spawn" below before you write anything.
If you want one snapshot of a whole world / level, use a `.scene`.
## Where it lives
- Source: `/zero/source/.../<Name>.bundle/`
- Identity: `<Name>` (the `.bundle` suffix strips for use in
identities, but the on-disk folder keeps the suffix).
- Folder shape (container type — open child set):
- `entity_template` — JSON of the entity hierarchy + components.
**Required.**
- `.metadata`: optional.
- Anything else: meshes / textures / materials / animations /
sub-bundles. Validator allows arbitrary additional files
(`allow_unlisted: true`).
## How to create one
Two paths, depending on whether you have an in-scene hierarchy or are
authoring from scratch:
```luau
-- From an existing in-scene entity: composes the given entity tree into a bundle.
local bundle = asset.create("bundle", "<Name>", { entity = entityRef })
-- Or scaffold an empty bundle, then update it from an entity:
local bundleRef = asset.create("bundle", "<Name>")
bundleRef:update(entityId)
```
## How to spawn
### 1. `Asset` component — live-linked (the canonical way)
**This is the primary way to put a bundle into a scene that survives
saving as an asset reference.** Spawn an empty entity, attach the
`Asset` component, point it at the bundle:
```luau
local car = entity.spawn("car_1")
car.component.add("Asset", { source = "<bundleIdentity>" })
```
What you get:
- The bundle's hierarchy spawns as **temporary** children under the
owning entity. They're real, addressable runtime entities, but
they are NOT serialized into `scene.json` on save, only the Asset component and its entity is saved.
- **Live editing of the bundle is supported.** When you edit
anything inside the `.bundle/` (entity_template, meshes,
materials, sub-bundles, components), every entity in the world
carrying that `Asset` component automatically tears down its
spawned children and rebuilds them from the new template — same
frame, no restart, no manual reload. The `onAssetReload` hook in
`Asset.component/init.luau` drives this.
- **Per-instance overrides survive across reloads.** Move a spawned
child, rename it, add/remove components on it, edit a component's
public state — the changes land in the Asset's `public.diff` (a
sparse delta keyed by the child's `original_id`) at save time and
are replayed by `bundleRef:instantiate()` on next load. The bundle stays
a single source of truth; per-spot tweaks ride on top.
- **Stable child ids across reloads.** `public.idMap` records the
`original_id → runtimeId` assignment on first spawn; on every
subsequent spawn the same mapping is reused so cross-entity
references (e.g. `SkinnedModel.skeletonRoot = "ent_xxx"`) stay
valid.
### 2. `bundleRef:instantiate` — baked, no live link
Resolve the bundle and instantiate it onto an entity — this spawns the
children **without** the live-link plumbing the `Asset` component
provides:
```luau
local car = entity.spawn("car_1")
asset.resolve("<Name>", "bundle"):instantiate(car)
```
What you get — and the trade-offs:
- The children spawn as **regular** (non-temporary) entities. On the
next `layers.active:save()`, the entire spawned hierarchy is serialized into
`scene.json` verbatim.
- **No live link to the bundle.** Editing the `.bundle/` after the
fact has no effect on already-spawned hierarchies. The bundle was
effectively consumed at spawn time.
- Scene-save / restore round-trip works the same as for any other
hand-authored entity tree.
Use this path when you want a one-shot expansion of the bundle into
permanent scene state (e.g. you're populating a scene's static
layout from a parametric bundle and don't want a live link).
You can convert a live-linked instance into a baked one explicitly
via `Asset:bakeIntoScene()` — this drops the `Asset` component and
promotes the temporary children to permanent scene entities.
## How it operates
1. **Registration.** Writing into a `.bundle/` folder indexes the
asset with the container-category restore plumbing.
2. **Inner refs.** Component fields that reference assets inside the
bundle use the `__ref` shape (`{ "__ref": "<guid>", "name": "...",
"type": "..." }`). The asset registry resolves them at spawn
time.
3. **Identity.** The `.bundle` suffix is stripped, so a bundle
resolves as `<Name>` (`asset.resolve("<Name>", "bundle")`); the
type argument tells it apart from any `<Name>` of another type.
4. **Hot reload.** Saving `entity_template` (or any file in the
bundle directory) fires the engine's per-asset content-changed
hook, which calls `onAssetReload("source")` on every entity
whose `Asset` component points at the bundle. Each carrier
despawns its children and re-instantiates against the new
template — typically next frame.
## Discovery
- `asset.list("bundle")` — every registered bundle.
- `asset.inspect("<name>")` — sub-asset catalog, source, this type
README.
- `ls /zero/source/<Name>.bundle` lists the files the folder holds.
## Parts and slots
A bundle can be an object assembled from parts, and a part that fills one. A
child carrying a `Slot` component is a named place another part fills (its
`kind`, the `tags` a part must carry, a `default` part, and where the part
attaches: that child's transform). A bundle whose root carries a `Part`
component is a part, and its `fits` names the slot kinds it fits. An
`Assembly` on the root holds what fills each slot by path (`wheel_fl`,
`engine/turbo`, a part's own slots nesting under the slot it fills), and
resolves, swaps and explains them at runtime:
```luau
local car = entity.spawn("car")
car.component.add("Asset", { source = "Buggy" }) -- a bundle with slots and an Assembly at its root
local asm = car.component.get("Assembly")
asm:configure({ wheel_fl = "SportWheel", engine = { part = "V8", turbo = "Turbo" } })
asm:fill("wheel_fl", "OffroadWheel") -- rebuilds that wheel alone
print(asm:fits("wheel_fl", "V8")) -- false, and why
```
The configuration is saved with the scene and synced; the parts are runtime
children rebuilt from it. `guides { path: "topics/assemblies" }` covers the
whole family.
## Authoring conventions
- Prefer `bundle:update()` or `asset.create("bundle", "<name>", { entity = ref })` over hand-authoring `entity_template`. The
in-scene composer wires `__ref` shapes correctly and pulls in every
transitively-needed asset.
- Keep bundles focused — one prefab per bundle. If you have a "vehicle
+ driver + sound" assembly, split into three bundles and reference
them from a scene rather than packing all three into one bundle.
- Name child entities meaningfully — bundle interiors are
inspectable, and stable IDs let other tooling (`entity_template`
edits, animation hooks) find them.
- Sub-bundles compose: a bundle's `entity_template` can `__ref` a
child `.bundle/`. The asset registry resolves them transitively.
## Common pitfalls
- **Confusing the two spawn paths.** `Asset` component = live link,
reference-only scene save. `bundleRef:instantiate()` directly = baked,
full hierarchy in `scene.json`.
- **Copying JSON between bundles.** A copied `entity_template` does not
bring the assets it references along with it; re-compose from a
live scene rather than copy-pasting JSON.
- **Live editing only fires for live-linked carriers.** Already-baked
entities don't update when you edit the source bundle.
## Related types
- `Asset.component` — the canonical mount point for a live-linked
bundle. See its README for the `idMap` / `diff` machinery in
detail.
- `.scene` — owns a whole world's worth of entities; spawns bundles
into itself via `Asset` components on per-spot entities.
## In the Inspector
Selecting a bundle opens its **Contents** section, from the bundle's inspect detail: the template's node count, its root, the component types attached across it, a table of each renderable node's mesh and material, and a link to each of those meshes and materials.
# yaml.module
YAML decode + encode for authored engine content. Pure Luau, no engine
dependencies — a general parsing library, not data-system plumbing.
```luau
local Yaml = require("@builtin::modules.yaml")
local doc = Yaml.decode(vfs.read(path))
vfs.write(path, Yaml.encode(doc))
```
## Supported subset
- Block mappings (`key: value`) and block sequences (`- item`), nested
to any depth.
- Compact sequence-of-mapping entries (`- path: x` / ` kind: y`).
- Flow collections — `{ a: 1, b: 2 }` and `[1, 2, 3]` — single line.
- Plain scalars with type inference: `~` / `null` → nil, `true` /
`false` → boolean, numeric literals → number, everything else →
string.
- Single- and double-quoted strings (`''` escapes a literal quote
inside a single-quoted string; `\n` `\t` `\r` `\\` `\"` inside a
double-quoted one). A quote opens a quoted scalar where a scalar
begins — after `key:`, after a `- ` entry marker, or inside a flow
collection — so a plain scalar carries an apostrophe as data
(`note: the world's rules`). Quoted keys are read in flow mappings.
- `#` comments — honored outside quotes, so a `#` inside a quoted
string is kept as data.
- Literal (`|`) and folded (`>`) block scalars, with a chomping
indicator (`-` strips the final newline, `+` keeps every trailing
blank line, none keeps one newline) and an indentation indicator
(`1`-`9`: the content's indent past the parent's), in either order
(`|-`, `>+`, `|2`, `>1-`). A block scalar's content is literal text:
quotes (balanced or not), `#`, `key: value` and `---` inside it are
data.
- An optional leading `---` document marker.
## Loud-error contract
Constructs outside the subset above raise instead of misparsing:
anchors/aliases (`&`, `*`), tags (`!`), directives (`%`),
multi-document streams (a second `---`), a block-scalar header
it cannot read (`|x`, `|-+`, `|0`), tab indentation, and inconsistent indentation (both block structure and
block-scalar bodies). Every error carries the offending line number
(`yaml: line N: ...`), so a bad config points straight at the line to
fix.
Tab indentation is rejected everywhere — including block-scalar body
lines whose leading whitespace contains a tab. A tab after the first
non-space character of a line is content and passes through.
## API
- `Yaml.decode(text: string): any` — decode a YAML document into a
Luau value (table, scalar, or nil for an empty document). Raises on
malformed input or an unsupported construct.
- `Yaml.decodeWithComments(text: string): (any, boolean)` — decode as
`decode` does, and report whether the text carries a comment by the
decoder's own rule: a `#` at a line's start or after whitespace, outside
a quoted scalar and outside a block scalar. Re-encoding the value drops
such comments, so an editor that writes a document back reads this first.
- `Yaml.encode(value: table): string` — encode a Luau table as YAML
(block style, two-space indent, alphabetically sorted keys). Raises
on values YAML can't represent (functions, userdata, non-string
mapping keys).
A top-level empty table encodes to an empty document, which decodes
back to `nil` (the YAML empty-document ambiguity).
# assetType.shared.settings
Resolves four questions about an asset's typed settings, all answered from its
type's `type.yaml` — nothing here is found by convention.
```lua
local Settings = require("@builtin::assetTypes.assetType.shared.settings")
-- or, from inside the assetType runtime:
local Settings = require("@builtin::assetTypes.assetType.shared").settings
```
## Where does a type's schema live?
`type.yaml` names it in a `settings:` block:
```yaml
settings:
values: "settings.yaml" # where an instance's own values live
schema: { type: "settings.luau" } # one schema for every asset of this type
```
or, for a type whose settings differ per instance (an importer):
```yaml
settings:
values: "settings.yaml"
schema: { instance: "settings.luau" }
```
or, for a type whose schema is a table its instance's own source declares and
whose values live elsewhere (a component: its `public` fields, valued on each
entity it is on):
```yaml
settings:
schema: { instance: "init.luau", export: "public" }
```
Whether an instance may lack its schema file is what the type's `files:`
block says of that file. Listed under `files.optional`, it may be absent, and
an instance without it declares no settings — the importer type lists it
there, since most importers take none. Listed under `files.required`, or not
listed at all, it must load, and an instance missing it is a broken
declaration and is logged.
`schema.type` is required at the folder relative to the TYPE's own
`.assetType/` directory (one schema shared by every instance).
`schema.instance` is required relative to each INSTANCE's own folder (`init`
resolves the instance's own identity; anything else is a sibling module under
it). Without `schema.export`, the file is required as a module and its return
value IS the schema. With `schema.export`, the file is run in an environment
of its own and `export` is read from a returned table carrying it, or else
from the globals the file assigned — a component's `public` is such a global.
That environment reads through to `_G` and takes every write itself; its own
`_G` is itself, its `declare` and `computed` record into it, and its `task`
records spawned, deferred and delayed work instead of starting it. Every other
global is the real one, so a call the file makes through one runs as it does
anywhere. The schema is the declared table's Field descriptors. A schema that
resolves to nothing says why through `Settings.schemaProblemOf(ref)`: the file
could not be read, compiled or run, or declared no such table. An instance
schema is read again when its file, its type's `type.yaml`, or the entry file
of any module the asset `require`s changes. `Settings.blockFor(typeName)`
returns the raw block; `Settings.schemaForType(typeName)` and
`Settings.schemaOfInstance(ref)` resolve the two schema shapes respectively;
`Settings.schemaFor(ref)` tries the type-level schema first, falling back to
the instance-level one. `ref` naming an assetType asset itself — e.g.
`asset.resolve("texture", "assetType")`, rather than an instance of it —
reports that type's own declared schema, the same one every instance of it
reports.
## Where do an asset's values live?
`settings.values` names a file living beside the asset's own content —
`Settings.valuesPathOf(ref)` joins it onto the asset's path.
`Settings.valuesOf(ref)` / `Settings.writeValues(ref, values)` read and write
it; this module only names where it is. A type that declares its schema with
no `settings.values` keeps no values in the asset — a component's live on each
entity it is on — so `valuesOf` answers nil for it, `asset.create` refuses a
`settings` table for it, and `asset.setSettings` refuses a write.
`Settings.recordedWrite(path, text)` writes a file the way every values file
is written, and an editor uses it for any other file it writes. It writes the
whole contents, records the write into the open undo operation when one is
open (so undoing it restores what the file held), and raises when the write
does not land.
## In what order are a schema's fields declared?
Every `Field.*` constructor stamps its descriptor with `declaredOrder`, one
more than the last. Luau evaluates a table constructor's entries in the order
they are written, so `Settings.orderedNames(schema)` lists a schema's fields —
and a struct's subfields — in the order the source writes them.
A values file, or a legacy `.metadata` settings block, may hold an older
spelling — a renamed field, an older enum member. Every read passes what is
stored through the type's `normalizeSettings` (`Settings.normalizeFor`), so
it answers in the fields the schema declares today, and writes nothing. The
next local edit of that asset rewrites the file in the declared spelling
(`Settings.healLegacyIfNeeded`).
`asset.setSettings(ref, patch)` writes a partial patch:
`Settings.resolvePatch(ref, patch)` normalises it, merges it over the current
values and applies the type's derived defaults for the fields it changed, and
`Settings.writeValues(ref, values, patch)` applies the corrections, checks the
schema and writes the file.
## How is a new asset's settings resolved at creation?
`asset.create` calls `Settings.resolveForCreate(typeName, name, destPath,
creationOpts, callerSettings, importerSettings)` before running the type's
`onCreate`, checks the result against the schema, writes it to the values
file, and hands `onCreate` a COPY of it as `opts.settings` (`Settings.
cloneValues`) — never the same table the values file gets, so nothing
`onCreate` does to its own copy can reach what was stamped. `onCreate` reads
settings from `opts.settings` and never writes to it. `asset.create` also
seeds the change baseline with the stamped values
(`Settings.seedChangeBaseline`), so the values file's own first dispatch
reports nothing changed and the payload `onCreate` encoded is the only
encode a create runs.
The layers, lowest priority first, each overriding the ones before it field
by field:
1. The schema's own declared defaults (`Settings.defaultsOf`).
2. The type's own `settingsDefaults(name, opts, asked)` hook, when its
`behavior.luau` exports one. `asked` holds what layers 3 to 5 name, so a
default can follow a field one of them sets.
3. The Preset Manager's matching row (`Settings.presetLayerFor`).
4. The creating importer's settings for this type.
5. The `settings` table the caller passed to `asset.create`.
6. The type's own `settingsDerivedDefaults(values, explicit, changed)` hook,
when its `behavior.luau` exports one — a field whose default follows
another field, set only when no layer from 2 to 5 named it.
7. The type's own `settingsCorrections(values, name, opts)` hook, when its
`behavior.luau` exports one — applied last, once every other layer has
merged.
Layers 2 to 5 are each read through the type's `normalizeSettings`, so a
layer naming a field in an older spelling sets the field it names today.
`asset.create`'s top-level settings alias accepts those older names too
(`Settings.acceptedFieldNames`).
Layer 7 is `Settings.applyCorrections(typeName, values, explicit, context)`,
the one corrections path every settings write takes — this create chain, and
`Settings.writeValues` for every later write (`asset.setSettings`, a type's
first-encode seed, the legacy heal). `explicit` names the fields the writer
asked for: the caller's `settings` table here, the patch for
`asset.setSettings`. A correction lands SILENTLY on a field the writer did not
name — nobody asked for that value, so overriding it just states a fact. A
correction that contradicts a field the writer named REFUSES the write,
naming the field, the value asked for, and why it cannot hold — the same as
every other place this codebase refuses an explicit request it cannot satisfy
(a `Field.checkSchema` violation, a BC7 alignment error) rather than silently
rewriting it into something else. So `asset.setSettings(tex, { compression =
"rgba16" })` on a mipped texture stores `generateMipmaps = false` beside it,
while `asset.setSettings(tex, { compression = "rgba16", generateMipmaps = true })`
is refused.
`Field.checkSchema` runs once, over the result AFTER the corrections hook has
applied — a bad value nothing corrects refuses the creation.
### Four hooks, four different jobs
Every hook is an optional export on a type's `behavior.luau`, alongside
`onCreate` / `onChange`. **None is ever handed a table to mutate** — each
reads what it's given and RETURNS a table of what it wants changed (or nil);
`resolveForCreate`, `resolvePatch`, `applyDerivedDefaults` and
`applyCorrections` are the only places that write to the values, by merging
what a hook returns the same way every other layer merges.
- **`normalizeSettings(values) -> values`** maps every older spelling a
stored file or a write may carry to the field and member the schema
declares now, returning a new table; a table naming both an older field
and its replacement keeps the replacement. The companion export
**`normalizedSettingNames = { ... }`** lists the older field names it
accepts beyond the schema, so a write naming one is not refused as an
undeclared field. `texture.assetType`'s maps the fields its earlier schema
stored to `colorSpace` + `compression`, and the older `filter` keywords to
`linear` / `nearest`.
- **`settingsDerivedDefaults(values, explicit, changed) -> partial?`**
answers the values of fields whose default follows another field. At
creation `changed` holds every field; on a write it holds the fields the
write moved. A returned field applies only where `explicit` — the fields
the writer named — does not name it.
- **`settingsDefaults(name, opts, asked) -> partial?`** supplies a DEFAULT
the schema itself has no syntax for — something only knowable at creation
time, like a naming convention, a fact read off a payload the caller handed
in, or a value that follows a field a higher layer sets (`asked` holds what
the Preset Manager, the importer and the caller name, the higher one's
value where two name a field). It runs early (layer 2 above), so any later
layer — Preset Manager, importer, caller — freely overrides it.
`texture.assetType`'s returns the `textureType` the name suggests (a
`_normal`-suffixed name is a `normalMap`), `compression = "uastc"` and
`maxDimension = 4096`, with `rgba8` for a `textureType` of `sprite` that a
higher layer names; for a pre-encoded ZTEX payload it returns the colour
space, compression and sampler the payload's own header carries. Either
way, an explicit `settings = { ... }` from the caller still wins.
- **`settingsCorrections(values, name, opts) -> (partial?, reasons?)`**
enforces a CROSS-FIELD CONSTRAINT the schema itself has no syntax for — a
fact about one field that depends on another field's resolved value, true
regardless of which layer produced that value. It runs last (layer 7
above), over the fully merged `values`, and again on every later write of
the values file. Nothing is left unmet in what gets
validated and stamped — but a value the caller explicitly asked for is
never silently rewritten to satisfy it; `applyCorrections` refuses instead
(see above). The optional second return, `{ [field]: reason }`, supplies
the "why" for that refusal message; a field with no reason falls back to a
generic one. `texture.assetType`'s returns `{ generateMipmaps = false,
maxDimension = 8192, colorSpace = "linear" }` whenever the RESOLVED
`compression` is high-precision (the encoder stores one verbatim, linear,
and refuses a mip chain or a downsample for it outright), paired with a
reason naming the compression, and `nil` otherwise.
## What does a declarer declare?
`Field.settingsOf(identity, mode)` embeds another declarer's schema as a
struct field without restating it. The identity names one of two kinds of
declarer, and `Settings.schemaOfDeclarer(identity)` tells them apart by what
`identity` resolves to:
- **An assetType** (e.g. `"@builtin::assetTypes.texture"`) — embeds the
schema every instance of that type carries, resolved via
`schemaForType`. This is the type-level shared schema from the section
above, not a per-instance one — an assetType has no settings of its own
to declare per instance.
- **Anything else** (an importer, or any other asset that declares settings
of its own) — embeds that asset's own schema, resolved via
`schemaOfInstance`.
Every schema this module serves has already run its `Field.settingsOf`
fields through this resolution — a caller reading `schema.someField.fields`
never sees an unresolved `declarer` with no `fields`. Resolution descends
into an already-literal struct's `fields` and a list's `element` the same
way, so a declarer embedded several fields deep is still reached. A declarer
that embeds itself, directly or through another declarer, stops one level
short of the loop instead of recursing forever.
## What are the defaults?
`Settings.defaultsOf(schema)` reads every field's declared `default` into a
plain table — a field declared without one is absent from the result rather
than holding a `nil` value.
# assetType.shared.typeYaml
A type's `type.yaml`, parsed. Every block a type declares in its `type.yaml`
(`settings:`, `inspector:`) is read through this module, so one file read and
one YAML decode serve all of them.
```lua
local TypeYaml = require("@builtin::assetTypes.assetType.shared.typeYaml")
local folder, identity = TypeYaml.folder("material")
-- "/zero/source/libs/@builtin/assetTypes/material.assetType", "@builtin::assetTypes.material"
local doc = TypeYaml.of("material")
print(doc.suffix) -- ".material"
print(doc.inspector.type) -- "inspector.luau"
```
- `TypeYaml.folder(typeName)` resolves the assetType and answers its folder
path and identity, or `(nil, nil)` when no such type resolves.
- `TypeYaml.of(typeName)` answers the decoded document, or nil when the type
does not resolve or its `type.yaml` does not decode to a table. The decode is
cached per type and read again whenever the file's content version moves, or
the type resolves to a different folder.
- `TypeYaml.described(typeName)` lists the files a type's assets must describe:
each file its `type.yaml` declares with a `described` content check, with the
keys and headings the check names and each key's byte limit. `TypeYaml.PLACEHOLDER`
is the marker a template puts where a description is yet to be written.
- `TypeYaml.describeBlock(typeName)` answers the `describe:` block of the
`type.yaml`: the type's own `guidance` on what a description covers and the
identities of `examples` whose descriptions show the shape.
# asset_change_dispatch (module)
Installs `_G.__zero_dispatch_asset_change`, the Luau half of the
asset-type **change-callback** system. The engine calls it once per VFS
source write (via `zero_scripting::ffi_callbacks::fire_asset_change_dispatch`,
queued as `VfsMutation::AssetSourceWritten`).
## What it does
1. Receives the written VFS path.
2. Walks up the path to the enclosing `<name>.<type>/` typed-asset folder
(deepest registered-type suffix wins, so the direct owner of the write
is chosen for nested typed assets).
3. Loads that type's `<type>.assetType/behavior.luau` via
`@builtin::assetTypes.assetType.shared.ref.loadTypeModule`.
4. If the module exports an `onChange` function, calls
`onChange(ref, change)` where `ref` is the typed `AssetRef` for the
changed asset and `change = { path, asset, type }`.
The engine runs each dispatch as its own yieldable task, so a hook can wait
on what it reads: a `vfs.read` of bytes only the blob store holds (a binary
file on the web build) fetches them. A hook that reads only what is resident
finishes inside the dispatch that started it. The writes routed to one asset
react in the order they were routed: a write routed while that asset's hook
is waiting queues behind it and reacts once the hook has finished.
This is the type-level analogue of the component-centric
`onAssetReload(field)` fan-out: components react to assets they
*reference*; an asset *type* reacts to writes inside its own *instances*.
## The source-write seam
A write inside a typed asset is also offered to the importer type's shared
layer (`onAssetWrite(assetPath, path, origin)`), the way a write with no
enclosing asset is offered to `onLooseWrite`. The asset's `import` record names
the file it was imported from (by guid and path); a write to that file re-runs
the importers on it, in place; the write that arrived is the only input. A type whose own `onChange` already keeps its retained source current
(a texture re-encoding its image, a sound clip its audio) sets
`reencodesRetainedSource = true` on its module, and the seam leaves that edit
to it.
## Re-entrancy
While an asset's `onChange` runs, a dispatch for that same asset made from
the hook's own coroutine is suppressed, so a handler that routes a write
back into its own folder inline doesn't recurse. A write the hook makes is
routed on a later frame like any other and reaches `onChange` again once
the hook has finished. `onChange` handlers must therefore be **convergent**: diff the meaningful
state before acting and short-circuit when there's nothing to do (the
`dynamicAsset` example only regenerates when the prompt actually changed
and bails while a generation is in flight). Idempotency is the contract,
exactly as it is for services' `start`/`stop`.
## Related
- `@builtin::assetTypes.assetType.shared.ref` — owns `loadTypeModule` + the per-type `ref` method
dispatch.
- `assetTypes/assetType.assetType/template/behavior.luau` — the documented
template that shows how to author `ref`, `global`, and `onChange`.
- `assetTypes/dynamicAsset.assetType` — the reference type that uses
`onChange` for prompt-driven regeneration with version control.
# asset_create
`asset.create` — the single, generic "instance a new asset of an existing type"
API. The same entry point scripts and agents use; there is deliberately no tool
wrapper, because agents author in Luau and call `asset.create(...)` directly.
```lua
local inst = asset.create("material", "my_metal", { base_color = { 0.8, 0.7, 0.2 } })
-- → AssetRef: inst.path == "/zero/source/my_metal.material", inst.guid, plus the
-- material type's ref methods.
```
`asset.create` makes a new *instance of an already-registered type* by running
that type's `onCreate(name, opts)` behaviour hook (declared in its
`behavior.luau`) and writing the produced files to
`/zero/source/<name>.<typeName>/` — the one authored location, in every mode.
While play runs, the play-mode write lock takes that write onto the **play
shadow**: live in the session, listed by
`vfs.playShadowPaths()`, and promoted or discarded on a guarded play-exit. The
asset created in play therefore carries the same identity, path and `require`
spelling it has in edit, and the session decides whether it stays.
A create whose output the world **reproduces** on every load — one made from a
component or a scene entrypoint — writes to the copy-on-write runtime store at
`/zero/runtime/assets/<identity>.<typeName>/` instead, so the code that rebuilds
it each load is its only source and the saved manifest never carries a second
copy.
## Placement
Four `opts` keys are consumed by the framework before the type's `onCreate`
hook runs, and steer where the instance lands:
| Key | Effect |
|-----|--------|
| `folder` | A relative subfolder under the source root: `/zero/source/<folder>/<name>.<typeName>/`. Groups a generator's output instead of accumulating it at the source root. |
| `into` | A resolved container ref (`.toolbox` / `.package`) to author INSIDE — lands at `<container>/<name>.<typeName>/` and registers as a member. Edit-mode only. |
| `dest` | An absolute destination path, owned by the caller (an importer building a `<name>.bundle/`). |
| `overwrite` | Re-author an existing destination in place, keeping its `.meta` guid so every reference stays valid. |
The `name` is always the asset's bare identity — the path goes in `folder`:
```lua
local rock = asset.create("mesh", "rock", { positions = p, indices = i, folder = "terrain/props" })
-- → rock.path == "/zero/source/terrain/props/rock.mesh"
```
`dest` and `into` both take precedence over `folder`. `folder` decides the
asset's identity, so it applies in every context: the runtime store is flat and
spells that identity in one folder name.
```lua
-- the same call from a component's awake()
-- → rock.path == "/zero/runtime/assets/terrain.props.rock.mesh"
-- → rock.identity == "terrain.props.rock", as it is from an execute
```
Types without an `onCreate` hook fall back to cloning their verbatim
`template/` skeleton to the same destination, so `create` works for every type.
Returns the created asset's `AssetRef` (`.path` / `.guid` plus the type's ref
methods — the same interned instance `asset.resolve` returns) on success and
raises (via `error`) on bad arguments or a failing `onCreate`.
Installed onto the FFI `asset` namespace by the prelude (`M.installInto(asset)`).
See `docs/specs/runtime-asset-copying.md`.
# declaration (module)
A type's `type.yaml`, decoded and held: the one reader behind every question
about a type as a whole — its files, whether it broadcasts its subassets, how
its instances are pictured.
## Exports
- `M.of(typeName) -> table?` — the decoded declaration, or nil when no type
answers to the name, it has no `type.yaml`, or the file does not decode (the
last with a warning naming the type). Shared: read it, do not write to it.
- `M.forget()` — drop every held declaration. The change dispatcher calls it
for every write inside an `.assetType` folder and for every removed type.
# asset_events
Per-asset runtime for declared asset events. An assetType's `behavior.luau`
declares an `events` schema the same way a component declares one; this
module turns that schema into the live objects an asset's ref fires and
listens on.
## Three faces, one Signal — keyed by the asset
The construction is `component_events`': one `Signal` per declared event,
reachable through a private fire-capable table, an owner-side emitter the
type's own behavior fires with, and a subscribe-only facade every other
holder of the ref sees. The facade exposes `connect` / `once` / `wait` and
has no `fire` at any key, so firing authority stays with the type.
What differs is the owner. A component event belongs to one instance on one
entity; an asset event belongs to the **asset**, so the runtime is keyed by
the asset's stable guid. Every resolver of that guid subscribes to the same
Signals, and a ref that is reclaimed and re-resolved re-attaches to the
subscriptions already there — the same reason `asset_ref`'s `runtime` table
is guid-keyed rather than stored on the envelope.
Both tables reject an undeclared event name: the private table's metatable
raises on a bad key and the facade raises on a bad key, so a typo surfaces
at the subscribe call instead of returning nil.
## Declaring events on a type
```lua
-- <name>.assetType/behavior.luau
local M = {}
M.events = {
changed = { payload = { value = Field.number(0, NoSync) } },
}
M.ref = {
poke = function(self)
local emitter = require("@builtin::assetTypes.assetType.shared.events").emitter(self.guid)
if emitter ~= nil then emitter.changed:fire({ value = 1 }) end
end,
}
return M
```
```lua
-- any holder of the ref
local ref = asset.resolve("@builtin::…")
ref.events.changed:connect(function(p) print(p.value) end)
```
## Exports
- `M.forAsset(guid, eventSchema)` — the runtime for one guid, built on first
use and reused after. A later call with a different schema reconciles:
surviving events keep their Signal and their subscribers, new events get a
fresh one, removed events are disconnected and dropped.
- `M.facade(guid)` — the subscribe-only view, or nil before a runtime exists.
- `M.emitter(guid)` — the owner-side `:fire` view, or nil before a runtime
exists.
- `M.has(guid)` — whether a guid holds a live runtime.
- `M.teardown(guid)` — drop one asset's runtime, disconnecting subscribers.
- `M.teardownAll()` — drop every runtime. Called on an engine mode flip so a
play-mode subscription does not survive into edit.
- `M.liveGuids()` — every guid holding a runtime, sorted.
## Notes
- Subscriptions made through the facade are ordinary `signal.module`
connections and disconnect the same way any other does.
- The runtimes table is held strongly, and its lifetime is bounded by the
mode flip that clears it.
# assetType.shared.inspector
Which inspector a type declares for its assets, and the declaration that file
returns. The editor's Inspector (`edui.inspector`) reads it to add, replace and
order the sections an asset shows.
## The block
A type names its inspectors in `type.yaml`, with either key or both:
```yaml
# the Inspector for every asset of the type, a file in the type's own folder
inspector: { type: "inspector.luau" }
# a file each asset may ship in its own folder, for what that asset declares
inspector: { instance: "inspector.luau" }
# both
inspector: { type: "inspector.luau", instance: "inspector.luau" }
```
`type:` is what selecting an asset of the type shows. `instance:` applies to
the things an asset declares, never to the asset's own view: a component
names `instance: "inspector.luau"`, and a component that ships that file
shapes its card on every entity it is on. Each value is a `.luau` file name.
A block that is not a map, names another key, names neither key or names a
file that is not `.luau` is malformed; the assetType validator reports it as
`assetType.inspector_block`, and a `type:` file missing from the type folder
as `assetType.inspector_file_missing`.
## The declaration
The file returns a table:
```luau
return {
-- sections this inspector adds (before Operations)
sections = {
{ id = "summary", title = "Summary", build = function(ctx)
return { ctx.widgets.text{ text = "mode is " .. tostring(ctx.settings.mode) } }
end },
},
-- a generic section id -> the section shown in its place
replace = { tags = { id = "tags", title = "Labels", build = function(ctx) return {} end } },
-- section ids, first to last; ids it does not name follow in default order
order = { "summary", "identity" },
}
```
A section is `{ id, title, icon?, open?, build(ctx) }`; `build` returns the
section's body as a list of edui nodes. Any member of the declaration other
than `sections`, `replace` and `order` must be a function (a helper the file
exports). `material.assetType/inspector.luau` is a full example.
The Inspector rebuilds an asset's sections when the asset's files change. A
section that shows state held outside those files names it with a
`signature(ref)` member: text that moves when that state moves, read every
time the Inspector checks for a rebuild, so it has to be cheap.
`renderFeature.assetType/inspector.luau` folds in whether the feature runs
and its live settings, so a feature enabled or reconfigured from a script
shows in the Inspector as it happens.
## API
```luau
local InspectorDecl = require("@builtin::assetTypes.assetType.shared.inspector")
InspectorDecl.blockFor("material") -- { type = "inspector.luau" }
InspectorDecl.blockProblem({ other = "x.luau" }) -- why a block is malformed, or nil
InspectorDecl.declaresTypeInspector("material") -- true: the asset view reads one
local decl, problem, file = InspectorDecl.declarationFor(matRef)
local cdecl = InspectorDecl.instanceDeclarationFor(compRef)
InspectorDecl.shapeProblem(decl) -- nil when well-formed
```
`declarationFor(ref)` (and `typeDeclaration(typeName)`) read the `type:` file:
- `(nil, nil, nil)` when the type names no `type:` inspector;
- `(nil, problem, file?)` when the block is malformed, or the file is missing,
raises while loading, or returns the wrong shape — the problem names the
file and what is wrong with it;
- `(declaration, nil, file)` otherwise.
`instanceDeclarationFor(declarerRef)` reads the `instance:` file in the
declaring asset's own folder, and answers `(nil, nil, file)` when that asset
ships none.
# scene_instantiable (module)
The instantiation contract's shared implementation. An asset type opts into
scene instantiation by defining `instantiate(self, target?, opts?)` on its
behaviour `ref` table; this module carries the parts that mean the same thing
for every type, so a caller writes ONE piece of code against all of them.
```lua
local root, idMap = ref:instantiate(target?, opts?)
```
## The contract
**IN — the base opts.** `target` is an owning entity ref: the instance lands
under (or, for a hierarchy type like a bundle, onto) that owner. With no
target the type spawns a fresh root. `position`, `rotation`, `scale`, `name`
and `temporary` place that root and mean the same for every type. `rotation`
takes three numbers as pitch/yaw/roll in DEGREES, or four as a quaternion. A
type may honour more opts of its own — `params` for a type built from declared
inputs, `idMap` / `diff` / `sourceTag` for the bundle override contract — and
says so in its own `instantiate` docs.
**OUT — the two returned values.** Every type returns the same pair:
- **`root`** — the composed root, as an `EntityRef`. Composition is
**synchronous**: the root and everything the type built under it are live
the moment the call returns, so a caller can parent to it, read its
components and hand it on in the same statement. There is no frame to wait
for and no callback.
- **`idMap`** — the `originalId -> runtimeId` map naming what the composition
spawned, `{}` for a type with no addressable children. **Never nil.** A
component that re-composes the asset on every load (`Asset`) keeps this map
and passes it back in, which is how a cross-entity reference into the
composition — `SkinnedModel.skeletonRoot` pointing at a bone — survives a
reload.
The return is enforced, not merely described: `AssetRef` dispatches every
`ref:instantiate(...)` through `result` on the way out, whether or not the
type called it. A type that returns something else fails at its own call
rather than handing its caller a nil root or a map that is sometimes absent.
## Exports
- `M.root(self, target?, opts?) -> root` — stand the root entity for a type
that spawns one: parented to `target`, born temporary when
`opts.temporary`, named `opts.name` else the asset's own name, placed.
- `M.place(root, opts?) -> root` — apply the placement opts to a root the type
already has (the adopt path: a bundle exploding onto its target, a
sceneModule reconciling under one).
- `M.result(self, root, idMap?) -> (root, idMap)` — return through the
contract. Checks `root` is a live entity ref, normalises a missing map to
`{}`, and errors naming the asset type when either is something else.
- `M.isOwned(opts?) -> boolean` — whether a component drives this call. The
`Asset` / `SceneModule` components tag their own calls with `sourceTag`;
they hold the reference and re-compose on every load. An untagged call came
straight from `ref:instantiate(...)` and has no such owner.
- `M.own(root, self, idMap?) -> root` — hand an already-composed root to an
`Asset` component pointing at `self`, so the reference and its map persist
and the composition is rebuilt on the next load. The component adopts the
live composition rather than building a second one.
- `M.check(value, constraint) -> ok, reason?` — the `sceneInstantiable`
field-constraint validator, also registered under that kind on load.
## The field constraint
`Field.instantiableRef` emits `constraint = { kind = "sceneInstantiable" }`,
and this module registers the validator for it on load. Given a written value:
1. `nil` passes — the field is optional (no asset assigned).
2. Reads the value's identity — a bare string, or a table's `__ref` / `guid` /
`identity` / `name`.
3. Resolves it (`asset.resolve(identity)` — any category) and asks the
resolved ref `ref:canInstantiate()`: true iff the asset's type defines an
`instantiate` method.
This is what lets `Field.instantiableRef` accept by CAPABILITY rather than a
hardcoded type list — a new scene-instantiable asset type is accepted the
moment it defines the hook, with no edit to the field or its consumers.
# asset_on_register
Drives the optional `onRegister(self)` assetType lifecycle hook — the surface that lets an asset instance run code exactly once when it first registers, operating on the per-instance ref (`self`). This is what makes "write an asset into the world → it takes effect live, no restart" work for content that registers into a runtime registry (editor panels, and anything else that opts in).
A type opts in by exporting `onRegister` from its `behavior.luau`. The hook fires exactly once per instance:
- On `engine.onWorldLoaded` — a batched sweep over every world instance of every type that implements `onRegister`, yielding every 32 instances so a large instance count never stalls a frame. The @builtin library (`/zero/source/libs/`) is excluded.
- On live `asset.create` — the create path calls `M.fireFor(ref)` on the freshly-created instance.
A per-VM once-guard keyed by the instance's VFS path ensures the two triggers can't double-register the same instance.
## Exports
- `M.fireFor(ref, onRegisterFn?) -> boolean` — fire one instance's `onRegister(self)` hook exactly once. Resolves the type's hook from `ref.type` when `onRegisterFn` isn't supplied. Idempotent: a second call on the same instance is a no-op. Returns true if the hook fired this call.
- `M.sweep()` — batched sweep: fire `onRegister` once for every world instance of every registered type that implements it. Yields, so it must run inside a task. Safe to call repeatedly.
- `M.install()` — install the world-loaded sweep. Idempotent. Runs the sweep on a system task so it ticks through pause and never blocks the caller.
# previewRecipe (module)
How a type makes its instances' pictures. A type declares it in its
`type.yaml`, under `preview:`, and every instance's `preview.png` is produced
by the recipe named there.
```yaml
preview:
kind: surface # instantiate | image | surface | custom
mesh: sphere # surface: the builtin mesh the asset is painted on
```
| `kind` | The picture is | Needs |
|---|---|---|
| `instantiate` | the asset stood in the studio through its own `instantiate` | — |
| `image` | a file the instance already holds, fitted to the frame | `source.file` or `source.files` |
| `surface` | the asset painted on a builtin mesh | `mesh` |
| `custom` | whatever the type's `behavior.luau` `ref.preview` returns | — |
`angle: [yaw, pitch]` (degrees) states the bearing the camera looks from, for
any kind that renders.
The `preview:` block says how the picture is PRODUCED. Whether search embeds
that picture is a separate declaration — an `image` chunk under
`indexing.content` naming `preview.png` — so a type states both, and a type
that states one without the other gets exactly what it stated.
## Exports
- `M.of(typeName) -> PreviewRecipe?` — the recipe `typeName` declares, or nil
when it declares none or no type answers to that name. Read once and held;
a block naming an unknown `kind`, or missing what its kind needs, is refused
with a warning naming the type.
- `M.forget()` — drop every held recipe. The change dispatcher calls it, with
`declaration.forget()`, for every write inside an `.assetType` folder and for
every removed type, so an edited `type.yaml` is read again.
# assetType.shared.ref
Builds the metatable every `AssetRef` (the `{ __ref, type, name, guid, identity, path }` envelope returned by `asset.resolve` / `asset.ref`) carries. Replaces the per-call C metatable that used to live in `crates/zero_scripting/src/ffi/bindings/asset.rs::attach_asset_ref_metatable`.
## Public API
```lua
local assetRef = require("@builtin::assetTypes.assetType.shared.ref")
-- Called by the Rust factory (push_asset_ref_handle) via
-- _G.__build_asset_ref_proxy after the envelope fields are set.
assetRef.build(envelope) -- attaches AssetRefMT, returns the table
```
The module exposes no invalidation surface — discovery is registry-driven and hot-reload is transparent (see "Discovery and hot-reload" below).
## Method dispatch order
`ref.foo` is resolved in this order:
1. **Default methods** — `getSource` / `getBytes` / `getText` / `exists` / `inspect` and the lazy `meta` property. Per-type definitions cannot override these (engine contract from gh#1889).
2. **Per-type `ref` table** — discovered through `asset.resolve(<typename>, "assetType")` + `require(<identity>.behavior)`. The lookup resolves to whichever `<typename>.assetType/` folder is currently registered for the typename (built-in, user, or library-vendored) and pulls its `ref` methods.
3. **Nil** — unknown key, same shape as the original C metatable.
## Adding methods for a new asset type
Drop a `behavior.luau` into the type folder:
```
src/lua/lib/assetTypes/<typename>.assetType/
├── type.yaml -- structural spec (existing)
├── behavior.luau -- ref + global behaviour (new)
└── template/ -- placeholder body (existing)
```
`behavior.luau` returns:
```lua
return {
ref = {
--!desc Per-instance method on every `<typename>` AssetRef.
myMethod = function(self, ...) ... end,
},
global = {
-- Reserved for the asset.<typename>.* surface — not consumed
-- yet; included so the contract from gh#1889 lands cleanly.
},
}
```
Within `ref` methods, `self` is the AssetRef envelope: `self.path`, `self.identity`, `self.guid`, `self.type`, `self.name` are available. Engine-required fields (`guid`, `path`, `identity`, `type`) must not be redefined.
## Discovery and hot-reload
Type discovery uses the asset registry, so it doesn't matter where a `<typename>.assetType/` folder lives — engine built-ins under `@builtin::assetTypes.<X>`, user types dropped under `/zero/source/<X>.assetType/`, or types vendored inside an imported library all resolve identically. The dispatcher walks `asset.resolve(<typename>, "assetType")` → canonical identity → `require <identity>.behavior` on every miss; the asset registry is the cache for the resolve half and Luau's `_LOADED` is the cache for the require half.
Hot-reload is transparent. When a `<typename>.assetType/behavior.luau` is rewritten through VFS, the engine's `invalidate_require_cache_for_state` re-runs the module and mutates the cached `_LOADED[<identity>.behavior]` table in place. Dispatch reads `mod.ref` on every access (rather than caching a sub-pointer), so existing AssetRefs see the new methods on the very next access — no Rust-side hook into this module is needed, and no invalidation API is exposed for callers to call.
The same registry-driven path also handles type **registration** correctness: dropping a new `<typename>.assetType/behavior.luau` updates the asset registry, the next `asset.resolve(<typename>, "assetType")` returns the new ref, and AssetRefs of that type immediately gain the new methods.
## Why Luau, not Rust
The asset surface needed per-category behaviour (`material:getProperties()`, `bundle:instantiate(entityId)`, `tool:run(args)`) and the old shape couldn't grow without an FFI change per method. Owning the proxy in Luau means:
- Type-specific methods ship as content (the `<typename>.assetType/` folder), not engine code.
- Adding a new asset type drops in a new folder; no Rust rebuild.
- Methods can call any other Luau global (`vfs`, `asset`, `Material`, `entity`, etc.) directly, instead of going through narrow FFI primitives.
## Consumers
- `crates/zero_scripting/src/ffi/bindings/asset.rs::attach_asset_ref_metatable` — the Rust factory that pushes envelopes invokes `_G.__build_asset_ref_proxy` once per push.
- Every Luau call site of `asset.resolve` / `asset.ref` / `entity(id).component.X.material` / any binding declared `AssetRef<...>` — they all receive envelopes built through this module.
# asset_ref_shapes
Publishes what every `AssetRef<category>` answers to, so a member read on an
asset-typed value is checked against the category's own surface.
An asset category's per-instance surface is authored: a
`<name>.assetType/behavior.luau` declares `M.ref = { ... }`, and each category
declares its own. The members an `AssetRef<inputMap>` carries are therefore
knowable only to `inputMap` itself — no fixed set of types covers the ones a
world defines, and the checker has no way to guess them.
This module reads each registered category's `M.ref` table from its source
(never executing it), renders it as a Luau table type, and hands the set to
the engine. From there a `ref:method(...)` on an asset-typed value resolves
against the category's real surface: a name it does not carry is reported
along with the ones it does.
## Results shaped by the asset
Some results are shaped by the asset rather than by its category —
`inputMapRef:activate()` answers one handle per binding THAT map declares, a
set that is authored and differs per map. A category states those by
declaring `refShapes`, and this module asks it once per instance and
publishes the answers keyed by asset identity. `types/assetType` documents
the authoring side.
## When it publishes
Publishing is driven by use: every entry point that produces diagnostics
calls `ensure`, which is a no-op once the set is current. A world load, a
write inside a type definition, and a write inside any asset whose container
computes shapes each mark it stale, so the cost lands on the next check and
only if one comes.
The sweep is complete and replaces what was published before, so a category
whose type is removed stops being published.
## Reading what the checker believes
From a call site, a type that is correct and one that was never published are
the same absence of a diagnostic. `published()` tells them apart:
```lua
local shapes = require("@builtin::assetTypes.assetType.shared.refShapes").published()
shapes.categories.inputMap
--> "{ activate: () -> any, controls: () -> { any }, ... }"
shapes.returns["@builtin::inputMaps.default"].activate
--> "{ crouch: Handle, interact: Handle, jump: Handle, ... }"
```
# texture
An **image asset** — the raw bytes of a `png` / `jpg` / `jpeg` / `hdr` /
`ktx2` / `webp` / `bmp` / `tga` file. A loose image written into the world is
imported into a `<stem>.texture/` container (see "Internal layout").
> New to how a texture exists as an asset, as CPU-side pixels, and as a live GPU
> texture (and how `renderer.texture.*` and render targets move between them)?
> Read `core/resource-model` first.
`texture` is a first-class assetType (not a bare plural-dir category) so
that:
- a texture has a stable asset **identity** and an explicit `asset_type`
link, and
- a material that references a texture records a real `.refs` edge to it
— which is what lets the CPU-residency system cascade from a material to
the textures it actually uses.
## CPU residency vs GPU residency
A texture can be resident two fundamentally different ways, and they must
not be conflated:
- **CPU residency** (this assetType): the raw image **bytes in memory** —
readable, hashable, decodable. You can operate on the bytes.
- **GPU residency**: those bytes uploaded to the device as a
GPU texture the renderer binds/samples. Byte-level reads are **not
possible** on the GPU variant — that is simply how the device works. GPU
residency is managed separately and is **not** part of this assetType.
Same texture identity, two separate residency states, two separate
operation sets, two separate reference counts.
## Internal layout
A `.texture/` holds these files, and its child set is open. `asset.create("texture",
name, ...)` and the **texture importer** (which creates through it when a loose
raster image lands in `/source/`) produce this shape:
```
<name>.texture/
source.<ext> the retained source raster — see below
source.ztex the full-resolution texels a cook dropped some of — see below
settings.yaml the texture's settings — see below
data.ztex canonical pre-encoded + pre-mipped payload (ZTEX magic)
.metadata what data.ztex holds and was encoded from
```
`asset.create` keeps a copy of the image bytes it was given as `source.<ext>`;
the importer moves the uploaded image in under its own name instead
(`foo.png` stays `foo.png`). Either way it is the texture's retained source:
rewriting it re-encodes `data.ztex` from the new pixels, and a rewrite carrying
the bytes `data.ztex` was already encoded from changes nothing.
`data.ztex` is the **preferred primary**: when present the renderer resolves
and uploads it verbatim — no re-mip. When absent the primary falls through to
a raster image in the folder, so a container holding only an image file reads
that image.
That fallthrough takes the lowest-named raster image in the folder and never
the rendered preview, which is made from the payload. A
container that lost its encoded payload and holds no other image has no
primary: `asset.diagnose(ref)` reports `noPrimaryFile`, `asset.validate(ref)`
reports `texture.no_payload`, and `renderer.texture.loadCpu` refuses it.
`asset.primaryFile(ref).path` names the file the resolution landed on; the
asset-system guide covers `diagnose`.
#### `settings.yaml` settings
A texture's settings are declared in `settings.luau` and live in its
`settings.yaml`. `asset.create` stamps them when the texture is made, and
`texRef:setSettings({ ... })` (or `asset.setSettings(ref, patch)`) changes
them. Editing an encode-relevant one re-runs the encode against the retained
source and rewrites `data.ztex`.
```lua
local tex = asset.create("texture", "hero_albedo", {
bytes = png, ext = "png",
settings = { maxDimension = 1024 },
})
tex:setSettings({ uastcRdo = "medium" })
```
| key | values | default | meaning |
|-----|--------|---------|---------|
| `textureType` | `color` / `normalMap` / `mask` / `sprite` | `color`, or guessed from the name | what the texture holds, recorded in the payload. `normalMap`: tangent-space normals — when `generateMipmaps` builds the chain, every level of it, the base included, is renormalised to unit length; with `generateMipmaps = false` the one stored level is the source as given, and the chain the GPU builds from it at upload is a plain box filter. `mask`: linear data (roughness, metallic, occlusion, height, masks). `sprite`: a 2D image with transparency. A new texture's type is guessed from its name — `_normal` / `_nrm` / `_n` / … → `normalMap`, `_rough` / `_metal` / `_ao` / `_height` / `_mask` / … → `mask` — unless it is given. |
| `colorSpace` | `srgb` / `linear` | follows `textureType`: `srgb` for `color` and `sprite`, `linear` for `normalMap` and `mask` | how the stored values are read when sampled. `srgb` for albedo/emissive/base-color, UI and sprites; `linear` for normal/RM/AO/height/mask data. |
| `compression` | `uastc` / `rgba8` / `rgba16` / `rgba32f` | a new texture: `uastc`, a `sprite` `rgba8` | what the texels are stored as — see "What each compression stores" below. |
| `effort` | `fastest` / `fast` / `thorough` | `fastest` | how long a `uastc` encode searches for each block — see "Block compression" below. |
| `uastcRdo` | `none` / `low` / `medium` / `high` | `none` | how far a `uastc` encode moves from the source to make a payload that compresses smaller on disk and over the network — see "Block compression" below. |
| `generateMipmaps` | `true` / `false` | `true` | bake the full mip chain into `data.ztex` at encode time. |
| `maxDimension` | integer, `1` to `8192` | a new texture: `4096` | the longest side the encoded texture may have; a larger source is downsampled to fit, and a `uastc` texture's longest side is the largest power of two within it. A value past 8192 is refused. |
| `filter` | `linear` / `nearest` | `linear` | sampler filtering recorded in the payload. A material slot and a UI image widget sample the texture with it. A post-process effect's texture property, a render pass's `@texture` input and a compute shader's `texture_2d` binding sample through a sampler of their own (the effect's `postprocess.setSampler`, the pass's, the compute shader's named `sampler` binding), which this setting does not change. |
| `wrapU` | `repeat` / `clamp` / `mirror` | follows `textureType`: `clamp` for `sprite`, else `repeat` | how a material samples the texture past its left and right edges: tiled, the edge texel held, or mirrored. Recorded in the payload. |
| `wrapV` | `repeat` / `clamp` / `mirror` | follows `textureType`: `clamp` for `sprite`, else `repeat` | the same, past its top and bottom edges. |
| `anisotropy` | `world` / `1` / `2` / `4` / `8` / `16` | `world` | samples taken along a surface seen at a grazing angle. `world` follows the `render.anisotropy` world setting (`renderer.setAnisotropy`); a number fixes it for this texture, clamped to what the device supports. A `nearest` texture samples with 1. A write may give the number itself (`anisotropy = 8`); it is stored as the string. Recorded in the payload. |
| `alpha` | `input` / `premultiplied` / `none` | `input` | what the encode does to alpha. `input`: kept as the source has it (straight alpha). `premultiplied`: colour multiplied by alpha — decoded to linear light, multiplied and re-encoded for an `srgb` texture, multiplied as stored for a `linear` one — the source a material with the `premultiplied` blend mode (`src + dst * (1 - src.a)`) composites. `none`: every texel opaque, colour kept. |
| `alphaIsTransparency` | `true` / `false` | follows `textureType`: `true` for `sprite`, else `false` | fill every fully transparent texel with the average colour of its non-transparent neighbours, spreading outward, so bilinear filtering and mip averaging never pull black into the edge of a transparent area. Alpha stays 0. Runs before `alpha` and before the mips. |
| `keepCpu` | `true` / `false` | `false` | keep the decoded pixels in the CPU store after the GPU upload — see below. |
The type's colour space, `alphaIsTransparency`, `wrapU` and `wrapV` are
stamped when the texture is created, and a later write that changes
`textureType` moves them with it, unless the same write names them.
A cook that drops texels — a `maxDimension` smaller than the source, or a
`uastc` cook resampling a side to a power of two — keeps the full-resolution
input as `source.ztex` (one RGBA8 level) when the container holds no source
image, so raising `maxDimension` later re-encodes from every texel. A cook
that drops nothing keeps nothing extra. A texture made from raw pixels whose
cook dropped nothing keeps its payload as its only source, so a settings
change re-encodes from that payload read back with straight alpha: what an
earlier encode discarded stays discarded, after `alpha = none` its alpha is
opaque, and a `uastc` payload re-encoded is the blocks' texels, not the
pixels first given.
A new texture's settings start from the texture type's own defaults —
`uastc` with a `maxDimension` of 4096; `rgba8` for a `sprite`; a 16-bit PNG
or `.hdr` source left to the schema's `rgba8` unless a layer names a
compression — and a world's Preset Manager rows
for `texture` override them (a world that ships at 2048, say), then an
importer's settings, then the `settings` the create names.
`filter`, `wrapU`, `wrapV` and `anisotropy` say how the texture is sampled,
not what its texels are. A write that changes only those rewrites the
payload's header (`renderer.texture.withSampler`): every level keeps its
bytes, and no preview is rendered, because no pixel changed. Every material
slot samples through its own texture's state, so two textures on one material
can wrap and filter differently. UI image widgets honour `filter` and draw each
image over its own bounds, so the wraps apply to materials. A post-process
effect, a render pass `@texture` input and a compute shader sample through
samplers of their own, so none of the four settings reaches them.
`keepCpu` is a lifecycle setting, not an encode input — it never re-runs the
encode. It acts on an already-materialised texture right away instead: `true`
loads the decoded pixels into the CPU store when the texture is GPU-resident,
`false` drops a resident CPU copy.
The `data.ztex` payload is a `ZTEX` magic, a 32-byte header and the mip
levels; the CPU encoder is exposed to Luau as
`renderer.texture.*` and consumed by the texture importer and this assetType.
#### What each compression stores
| `compression` | In `data.ztex` | On the GPU | Limits |
|---|---|---|---|
| `uastc` | UASTC 4x4 blocks, whole mip chain, one byte per texel, each level compressed for storage. | Transcoded at upload to a block format the adapter samples: `Bc7RgbaUnorm[Srgb]` where it samples BC textures (desktop native, desktop WebGPU), ASTC 4x4 where it samples ASTC and not BC (phones, a browser on one), ETC2 RGBA8 where it samples only ETC2, each one byte per texel; RGBA8 at four bytes per texel on an adapter that samples none of them, and for a texture with a side under four texels. | each side is resampled to the nearest power of two (halfway rounding up), nearest-neighbour for a `nearest` texture. |
| `rgba8` | From an image with `generateMipmaps = false`, a `maxDimension` the image fits within, `alpha = input` and `alphaIsTransparency = false`: the source PNG / JPEG / WebP itself, decoded at load. Otherwise raw 8-bit RGBA with the encoder's box-filtered mip chain. | `Rgba8Unorm` / `Rgba8UnormSrgb`, four bytes per texel. | none |
| `rgba16` | Raw 16-bit unsigned normalized, one level. An image is decoded at its own depth: a 16-bit PNG keeps every bit, an 8-bit channel `v` is stored as `v * 257`. The frames of an animated image are decoded at 8 bits. | `Rgba16Unorm` where the adapter grants 16-bit normalized textures (native); half float (`Rgba16Float`, about 11 significant bits) where it does not (every browser). | one level, no `maxDimension`, `linear` only. |
| `rgba32f` | Raw 32-bit float, one level. An image channel is stored as its value over the channel's maximum. | `Rgba32Float`. Where the GPU filters 32-bit float, every consumer samples it. Where it does not (every browser): a material slot draws the shader's declared default; a UI image draws nothing; a render pass naming it as an `@texture` input is skipped; a compute shader that declares no sampler binds it and reads it with `textureLoad`, and one that declares a sampler refuses the dispatch — each logging once, naming the texture and its format. CPU reads (`:load()`, `renderer.texture.decode`) return every bit on every device. | one level, no `maxDimension`, `linear` only. |
A high-precision compression turns `generateMipmaps` off, `maxDimension` to
8192 and the colour space to `linear` when the same write does not name them,
and refuses a write that names a contradicting value.
#### Block compression
`compression = "uastc"` stores every mip level as UASTC blocks — sixteen bytes
per 4x4 texels — a format no GPU samples directly but every one can be handed:
at upload each level is transcoded to a block format the adapter samples, BC7
on a desktop GPU and ASTC 4x4 on a phone's (ETC2 on one that samples ETC2
alone), so one payload is resident at one byte per texel on each, and RGBA8 on
an adapter that samples none; `renderer.textureMemory().blockFormat` names the
format this adapter holds it in. Only the levels a device keeps are
transcoded: under a texture-size cap (`renderer.setMaxTextureSize`) a texture
is uploaded from the first level of its chain within the cap. Compression is
lossy, which makes it the choice for colour, normal and detail maps; `rgba8`
holds a texture exactly, which is what pixel art and a mask whose exact codes
matter need.
`effort` is how long the encode searches for each block. `fastest` tries one
candidate for an opaque block, `fast` takes about 6 times as long and
`thorough` about 17 times, each landing closer to the source texels. What the
GPU holds is the same size at every effort, and so is the time a load takes.
`uastcRdo` trades accuracy for a payload that compresses smaller: `none`
encodes each block as close to the source as the encoder finds, and `low`,
`medium` and `high` let a block move further from the source when a nearby
block's bytes would let the payload compress better. What the GPU holds is the
same size either way; what moves is the size of `data.ztex` on disk and over
the network, and the encode time, which a pass more than doubles.
The encode runs on the CPU when the texture is created or a setting that
decides its texels changes. A call from a task that can yield cooks off the
frame and the task waits for it, so the world keeps running: a 1024x1024
texture with its mip chain takes about a second on a desktop CPU.
A texture whose settings say `rgba8` is never compressed.
## Reading an asset from its ref
`texRef:settings()` reads the container's settings with every schema default
filled in and `texRef:setSettings({ maxDimension = 1024 })` patches them;
`texRef:load()` decodes the payload's pixels into the CPU store for per-pixel
access, `texRef:handle()` puts the texture on the GPU under the asset's own
guid, and `texRef:preview()` renders it as a swatch.
`texRef:info()` reads the compiled `data.ztex` payload's header: the format it
was compiled to — also as its `compression` and `colorSpace` — the
`textureType`, `alpha` and `alphaIsTransparency` it was encoded under, its width and height, its mip and array-layer counts, the
sampler recorded in it (`filter`, `wrapU`, `wrapV`, `anisotropy`), and, for an animated sequence, each layer's
display time and the length of one pass:
```lua
local info = texRef:info()
print(info.width, info.height, info.format, info.mipCount)
```
It answers off that payload's own header, so a compiled texture reports how big
it is for a header read where `load()` decodes its pixels and `handle()` uploads
them. A container whose primary is a source image answers `(nil, err)`, and
`load()` reads its size. The `compression` and `colorSpace` it reports are what
the payload holds; the settings table holds what the encode was asked for.
## In the Inspector
Selecting a texture opens its **Texture** section right below the asset's
identity, then its Settings. The section draws the encoded texture through the
texture's own sampler, so a `filter = "nearest"` texture reads crisp at every
zoom. The zoom row offers **Fit** and each integer zoom (1×, 2×, 4×, 8×) whose
drawn size fits 512 px; a 128 px texture offers up to 4×, and one wider than
512 px offers Fit alone. Below the view, the section lists every field of the
`data.ztex` header (`texRef:info()`) — size, format, mip levels, filter, layers,
and a sequence's frames and duration — and the GPU memory the encoded mip chain
takes across its layers. The filter and the other encode settings change in
the Settings section; the Texture section reports the re-encoded result.
# `.font` asset type
A `.font` is a first-class font asset: font-file bytes (TTF / OTF / WOFF /
WOFF2) converted into a guid-anchored asset via
`asset.create("font", name, { bytes = … })`.
A font is a **general CPU resource** usable by any text surface. The
`onRegister` hook calls the engine `font.register(name, bytes)` primitive,
which installs the parsed face into **both** the 2D/3D text system (`text.*`)
and the egui UI text system, under the family name the font file itself
reports. A registration is what puts the face in the database; whether
`fontFamily = "<name>"` selects it is the separate question of whether the
shaper matches that name, and `font.reconcile()` answers it per family:
```lua
for _, f in ipairs(font.reconcile()) do
if not f.selectable then
print(f.family .. " is registered but shapes as " .. f.shapedWith)
end
end
```
`text.face(handle)` gives the same answer for one label. A family the shaper's
database holds is selectable under its own weights and for the glyphs its faces
cover: a row that reads `familyCoveredNoGlyph` names a family the shaper picks
whose faces have no glyph for the text asked of it. `topics/text` covers reading
a label that came out in the wrong face.
Fonts are CPU resources (not GPU), so the primitive lives in the `font.*`
namespace, the CPU analogue of `renderer.texture.create`.
## Layout
```
<name>.font/
data.zfnt the baked payload `asset.create("font", ...)` writes
face.json optional: family group, role, aliases, on-demand entries
```
## Editing a font's files
A write to a file the registration reads (`face.json`, `data.zfnt`, or a raw
font file in the folder) installs the font again from what the folder now
holds. The names the new `face.json` gives it (its group and each alias) select
it on every text surface from then on, and a name the earlier descriptor gave it
and the new one does not stops selecting it; a changed `group` moves the face to
that group. Text laid out under a name the edit decides is laid out again, and a
`Text3D` label draws it again on the next frame. For an on-demand face the
same holds for the font and coverage files its entries name: the entries are
declared again from their coverage files, and an entry already installed is
registered again from its font file with the names the descriptor now gives.
## On-demand faces
A face whose file is large and whose script most sessions never draw is
declared by the codepoints it covers instead of being read at boot. Its
`face.json` lists the font files of the folder under `onDemand`, each beside a
coverage file:
```
NotoSansCJK.font/
face.json { group, role, aliases, license, onDemand = [ { name, file, coverage }, ... ] }
kana.otf a font file (TTF / OTF)
kana.coverage the codepoints it maps: "3000-30ff,ff01-ff5e,..."
...
OFL.txt the license the files are distributed under
```
`onRegister` reads the coverage files and calls `font.declare(name, coverage,
{ file, family, role, aliases })` for each entry; no font file is read. The first
time a label or a `text.*` object meets a character an entry covers and no
installed face draws, the engine requests that entry: its file is read without
blocking the frame and registered with `font.register`, and from then on every
text surface draws the character from it. `font.onDemand()` reports each entry
as `declared`, `requested`, `installed` or `failed`, and `font.prefetch(text)`
requests the entries a text needs before it is drawn.
`font.coverage(bytes)` writes a coverage file from a font file's character map.
Splitting one large face into several entries (by script, and by how common a
character is) means a session reads only the part its text reaches: the
built-in `NotoSansCJK.font` is cut that way, and a page of Japanese reads its
kana and common-ideograph entries alone.
Every registered face, on-demand or not, is a per-glyph fallback for every
family the UI holds: a character the selected family has no glyph for is drawn
from the first face that has one.
## In the Inspector
Selecting a font opens its **Font** section: the family the face registers under, its role and aliases from `face.json`, whether the text system has the family registered and, when it holds it, how many faces and which weights. Below the facts, a sample line is drawn in the family at 14, 20 and 32 px.
# Shader (asset type)
A shader is a WGSL program that runs on the GPU. Materials reference shaders and
supply property values; renderable entities reference materials, not shaders
directly. This README is the canonical reference for writing **every** kind of
shader — it covers all domains, the surface authoring contract, the material
property schema, the engine library functions you can call, and how a shader
registers and compiles.
> New to how shaders, materials, meshes, and textures relate as assets and live
> GPU resources? Read `core/resource-model` first.
## Folder shape
A `.shader` is a folder asset:
```
<name>.shader/
shader.wgsl # the WGSL body (required)
properties.yaml # the material property schema (optional; surface shaders)
README.md # a one-paragraph description of THIS shader (required)
.metadata # asset metadata
```
Each file carries a committed `.meta` sidecar pinning its stable guid. The
identity is `<name>` (the `.shader` suffix strips).
A shader's material properties are exactly the ones its `properties.yaml`
declares: a shader with no `properties.yaml` declares none, and the engine
reads no property from the WGSL body. A `.wgsl` file standing on its own,
outside a `.shader` folder, is WGSL source rather than a shader: a material
draws with the `.shader` folder that holds it as `shader.wgsl`. Materials and other assets
reference a shader by **AssetRef** (a typed, guid-backed handle), never by a raw
string.
## Domains
The first line of `shader.wgsl` declares the domain:
```wgsl
// @domain: surface
```
| Domain | What it is | You write |
|---|---|---|
| `surface` | A material on a mesh (the common case). | `fragment()` (+ optional `vertex()`/`vertex_clip()`). The engine generates everything else. |
| `post-process` | A full-screen pass over the rendered frame, in the space it declares (bloom, vignette, grade). | `fragment(in: PostInput) -> vec4` only. The engine generates the fullscreen `vs_main`/`fs_main` and the group(1) interface from `properties.yaml`. |
| `screen` | A standalone full-screen visual (UI background, procedural). | `fragment(in: ScreenInput) -> vec4` only. The engine generates the fullscreen `vs_main`/`fs_main` and the group(1) interface from `properties.yaml`. |
| `sky` | The sky background (a fullscreen material). | `fragment(in: SkyInput) -> vec4` only. The engine generates the fullscreen `vs_main`/`fs_main` and the group(1) interface from `properties.yaml`. |
| `compute` | GPU data processing, no rasterization. | An `@compute @workgroup_size(...)` entry over storage buffers; you write the `@group/@binding` lines yourself. For compute work, the `.computeShader` assetType (see below) generates them for you. |
`// @domain:` resolves the names above and `postprocess` / `screenspace` / `ui`
as further spellings of `post-process` and `screen`. A body carrying no
`@domain:` line is `surface`. `shaderRef:domain()` answers the domain a body
declares, a further spelling as the domain it names. Any other name is refused where the shader
registers, with an error naming the spellings that resolve, so a shader reaches
the renderer only under a domain the compiler scaffolds.
Particles are drawn and simulated through these domains like anything else: the
particle package's `particle_sprite` and `particle_mesh` are `surface` shaders,
and its simulation, culling and sorting steps are `.computeShader` assets.
## Texture coordinate space
A shader that samples a texture through a coordinate it builds itself declares
which space that coordinate lives in, on a line beside `@domain:`:
```wgsl
// @domain: surface
// @uv_space: world
```
| Value | The coordinate | A material's `uv_scale` means |
|---|---|---|
| `mesh` (the default) | The mesh's own UV attribute, `input.uv`. | How many times the texture repeats across a surface. |
| `world` | The fragment's world position, as a triplanar projection samples. | How many times the texture repeats per world unit, so a larger surface carries more. |
Texture residency reads the declaration to work out how finely a frame samples
each texture: a material laying eight copies of its texture across a surface
samples each copy at an eighth of the surface's size, and the mip level that
serves it is three levels coarser than the surface's own footprint implies.
Leaving the line out puts the shader in mesh space, which is what sampling
`input.uv` is. `renderer.textureStreaming().textures` reads back the span each
texture was measured at and the level that span asked for.
**Compute shaders have their own assetType.** A `.computeShader/` folder gives compute the same zero-scaffolding contract surface/sky/post/screen get: you write only `@compute fn main` (+ helper fns/structs, **no `@group/@binding` lines**) and declare bindings in a sibling `bindings.yaml` (storage buffers, textures, samplers, storage textures, plus a `params:` uniform); the engine generates the `@group(0)/@binding(N)` declarations. Dispatch it through a resolved reference (`asset.resolve("name", "computeShader"):dispatch(...)`). See the `.computeShader` assetType README and `man compute`.
**Surface**, **sky**, **post-process**, and **screen** shaders all get
engine-generated scaffolding — you write only `fragment()` and declare properties in
`properties.yaml`; the engine generates the entry points and the group(1) material
interface. (Each domain's `fragment` takes a domain-specific input instead of
`FragmentData`: `SkyInput`, `PostInput`, or `ScreenInput` — see "Sky shaders" and
"Post-process & screen shaders" below.) Compute shaders get the same treatment via
the separate `.computeShader` assetType (write only `@compute fn main`, declare
bindings in `bindings.yaml`). The `compute` domain here writes its own
entry points directly — see `man compute`.
## Surface shaders
A surface shader writes **only** the parts that are unique to it. The engine owns
all the scaffolding — the vertex stage, every render-mode entry point
(`vs_main` / `fs_forward` / `fs_gbuffer`), group(0) (camera, transforms, lights,
shadows, and the material rows), and the material interface generated from
`properties.yaml` (`material` and your texture slots). Adding a render path never
touches your shader.
Inside `fragment()` you have access to exactly four things — nothing is hidden:
1. **`input: FragmentData`** — the interpolated per-fragment surface data the
engine hands you: world position, world normal, uv, view direction, vertex
colour, and more. Every field is in the `FragmentData` table below.
2. **`material.*` and your textures** — the material inputs **you** declare in
`properties.yaml`. The mapping is mechanical: a scalar property named `foo`
becomes `material.foo`; a texture property named `bar` becomes `bar` (the
texture) and `bar_sampler` (its sampler). `material` has **exactly** the fields
your `properties.yaml` declares and no others — to add an input, add a property.
List what a shader exposes with `shaderRef:getProperties()` or
`asset.inspect("<name>")`.
3. **Engine state** — the scene's camera, lights, time, and shadows that the
renderer binds for every surface shader. `uniforms.*` (sun direction/colour,
ambient, `camera_pos`, `time`) and the punctual lights (`zero_point_light(i)`)
are always in scope; the shadow functions come from `#include "@builtin::shaderModules.shadows"`. This is how
you light a surface **without** the prebuilt PBR — see "Engine state" and
"Writing your own shading" below.
4. **Library functions** you `#include` (see "Library includes").
A shader and its property schema go together. The two files below are a complete
surface shader:
`properties.yaml` — declares the material inputs:
```yaml
properties:
- { name: base_color, type: color, default: [1, 1, 1, 1] }
- { name: roughness, type: float, default: 0.5, min: 0, max: 1 }
- { name: metallic, type: float, default: 0.0, min: 0, max: 1 }
- { name: base_color_texture, type: texture, default: white }
```
`shader.wgsl`; those declarations are read as `material.base_color`,
`material.roughness`, `material.metallic`, and the `base_color_texture` /
`base_color_texture_sampler` pair:
```wgsl
// @domain: surface
#include "@builtin::shaderModules.pbr_shading" // the PBR helpers — see "Library includes"
fn fragment(input: FragmentData) -> vec4<f32> {
var s = zero_pbr_default_surface(input);
s.albedo = material.base_color.rgb
* textureSample(base_color_texture, base_color_texture_sampler, input.uv).rgb;
s.perceptual_roughness = material.roughness;
s.metallic = material.metallic;
return vec4<f32>(zero_pbr_shade(input, s), material.base_color.a);
}
```
`fragment()` returns the final colour. The engine forces no lighting model on
you — flat, toon, PS1, dithered, or PBR are all just different `fragment()`
bodies. Calling `zero_pbr_shade` is opt-in.
### The two surface bodies, and what each gives up
A surface shader exposes one of two entry points, and which one decides how
much of the shader the engine can see:
| Body | The engine gets | PBR debug passes read |
|---|---|---|
| `fn fragment(input: FragmentData) -> vec4<f32>` | the finished pixel, and the `PbrSurface` too when the body hands one to `zero_pbr_shade` | the real channel where a surface was handed over; magenta, the "no data here" marker, where the body shades by its own math |
| `fn surface(input: FragmentData) -> PbrSurface` | the surface (albedo, roughness, metallic, emissive, normal, occlusion), which the engine then lights | the real channel |
Both are first-class: `fragment()` is what you want when the shader owns its
own look, `surface()` when you want engine lighting. `capture { pass = "albedo" }`
(and `roughness` / `metallic` / `ao` / `emissive`) answers about your material
for either body wherever the engine has the surface — writing the same
`PbrSurface` behind `surface()` or behind `fragment()` + `zero_pbr_shade` gives
one answer per channel. A body that lights itself by its own math draws those
channels magenta. `pbr.shader` exposes `surface()`.
`capture { pass = "normal" }` is a geometry read rather than a material one, so
it answers for every shader: the normal of the surface where one was handed
over, and the interpolated normal of the triangle where none was. It is the same
normal the frame's normal buffer carries, which is what a screen-space pass
reading `@scene.normal` sees.
`shaderRef:shadingModel()` reports which one a shader ended up with —
`"engine-lit"` for `surface()`, `"self-shading"` for `fragment()`.
### Optional vertex hooks
```wgsl
// Displace geometry in OBJECT space. Runs on skinned meshes too (it sees the
// already-skinned, morph-applied local position). Return the modified vertex.
// The `v` parameter is immutable (WGSL function params are), so copy it into a
// `var` first and modify that.
fn vertex(v: VertexData) -> VertexData {
var out = v;
out.position += out.normal * sin(out.time) * 0.1;
return out;
}
// Post-transform vertex hook (runs AFTER the engine's world/skin/projection). Two
// things you can do here, separately or together:
// * move `clip_position` — vertex jitter, PS1 snapping, render-resolution lowering;
// * write `color` — PER-VERTEX (Gouraud) shading: light the vertex from its
// world_normal + the light context and the engine forwards the result into the
// interpolated `vertex_color` the fragment reads.
fn vertex_clip(v: ClipVertex) -> ClipVertex {
var out = v;
// Gouraud: Lambert sun + ambient, evaluated once per vertex (flat-banded shading).
let n_dot_l = max(dot(v.world_normal, -normalize(v.light_direction)), 0.0);
let lit = v.ambient_color * v.ambient_intensity + v.light_color * v.light_intensity * n_dot_l;
out.color = vec4<f32>(v.color.rgb * lit, v.color.a);
out.clip_position = round(out.clip_position * 80.0) / 80.0; // PS1 snap
return out;
}
```
#### Geometry built per instance: grass, kelp, hair cards, debris
`vertex()` knows which vertex of which instance it places:
`VertexData.vertex_index` is the vertex's index in its mesh and
`VertexData.instance_id` the instance's slot, whose lanes
`zero_object_data(v.instance_id, lane)` reads. A hook can therefore build its
geometry rather than displace it: a strip mesh drawn thousands of times by
`renderer.mesh.drawInstanced`, each copy shaped, sized and bent from the lanes
of its `instanceDataBuffer`, becomes a field of grass blades, kelp fronds or
hair cards that sway with `v.time`. Being a surface shader, it is lit,
shadowed (it casts the shape the hook built), motion-vectored and fogged like
the rest of the scene. The rendering-instance-data guide
(`guides { path = "topics/rendering-instance-data" }`) has a worked blade
field; a render feature's raster pass (`types/rasterShader`) is the
hand-written alternative, a vertex and fragment program with its own pipeline.
#### Declaring how far a hook moves geometry: `vertexBoundsDilation`
Static geometry is culled per cluster against its bounds as authored. A hook that moves vertices — wind sway, an outline hull, a vertex wave — can move them outside those bounds, and the cull then drops clusters whose displaced triangles are still on screen. Declare the farthest the hook moves any vertex, in world units, as a `vertexBoundsDilation` float property, with a default that covers the shader's own properties across their ranges:
```yaml
properties:
- { name: sway, type: float, default: 0.5, min: 0, max: 4 }
# 0.4 x sway, at sway's largest value.
- { name: vertexBoundsDilation, type: float, default: 1.6, min: 0, max: 100 }
```
Each material sets its own value — a material that sways less can set less, one moved farther sets more — and a material that sets none takes the default. The cull adds the material's value to every bound in its frustum and occlusion tests (not to the level-of-detail cut). A shader with a hook and no `vertexBoundsDilation` property is named once in the engine log when it compiles; its geometry is culled against its undisplaced bounds.
### Structs the engine passes you
`FragmentData` (input to `fragment()`):
| Field | Type | Meaning |
|---|---|---|
| `world_position` | `vec3<f32>` | fragment world-space position |
| `world_normal` | `vec3<f32>` | interpolated geometric normal (normalized) |
| `world_tangent` | `vec4<f32>` | xyz = world tangent, w = handedness (mikktspace) |
| `uv` | `vec2<f32>` | primary texture coordinates (perspective-correct) |
| `affine_uv` | `vec2<f32>` | the same UVs interpolated **without** perspective correction — the PS1 texture "swim". Always available; sample with it instead of `uv` for affine texture mapping |
| `vertex_color` | `vec4<f32>` | interpolated per-vertex colour (white if unset; carries per-vertex lighting if `vertex_clip()` wrote it — see Gouraud below) |
| `view_position` | `vec3<f32>` | camera world position |
| `view_direction` | `vec3<f32>` | normalized world-space direction the camera sees this point from: toward the camera's position under a perspective projection, against its forward axis (the same for every point) under an orthographic one |
| `light_direction` | `vec3<f32>` | directional (sun) travel direction, unit length; surface→sun is `-light_direction`. Where no sun reaches the camera it points straight down |
| `light_color` | `vec3<f32>` | sun colour |
| `light_intensity` | `f32` | sun intensity; 0 where no sun reaches the camera |
| `ambient_color` | `vec3<f32>` | ambient colour |
| `ambient_intensity` | `f32` | ambient intensity as it reaches this pixel: the scene's ambient occlusion is already in it (times `ambient_visibility`), so a crease darkens under the ambient with no work in the shader |
| `ambient_visibility` | `f32` | the share of the light of its surroundings that reaches this pixel: 1 minus the scene's ambient occlusion here, 1 where nothing occludes it. Multiply it into any light of the surroundings you gather yourself (a sky colour, a reflection-probe or environment sample); `ambient_intensity` already carries it. `zero_scene_bent_normal(input.frag_coord.xy, n)` names the direction the open part of the hemisphere faces, `n` where no pass wrote one |
| `frag_coord` | `vec4<f32>` | framebuffer pixel coords (xy) + depth (z) — for screen-space dither |
| `receives_shadows` | `f32` | per-instance: > 0.5 if this surface receives shadows |
| `material_index` | `u32` | per-instance: the index standing for the material this surface draws with — the number `renderer.materialIdentity()` maps to a name |
| `shader_data` | `array<vec4<f32>, 4>` | per-instance: this entity's own four `vec4` lanes — its `ecs.ShaderData` component's `lane0`..`lane3`, or what `renderer.instanceData.set` put in them (zero until something does). A procedural graph's elements set them through the `data0`..`data3` attributes |
| `front_facing` | `bool` | true for front faces (double-sided shading) |
| `time` | `f32` | gameplay seconds: advances with the world at `engine.timeScale` and stands still while gameplay is paused (edit mode starts paused), so an effect slows and stops with the world |
`shader_data` is the per-instance channel, and it is what lets ONE material serve
many entities that differ in a value — a dissolve at its own progress per subject,
a hit flash at its own age, a fill level per instance. The alternative is a
material per entity, which is a separate pipeline binding and a separate row in
the material table for every one of them.
```wgsl
const DISSOLVE_LANE: u32 = 0u;
fn fragment(input: FragmentData) -> vec4<f32> {
let progress = input.shader_data[DISSOLVE_LANE].x;
...
}
```
```lua
-- one material on twenty entities, each at its own progress
renderer.instanceData.set(subject, 0, progress)
renderer.instanceData.clear(subject) -- back to zero lanes
```
The engine attaches no meaning to a lane: a shader and whatever writes it agree
what each one carries. Name the indices in the module that writes them.
`renderer.instanceData.laneCount()` is how many there are. The same lanes reach a
hardware-instanced population through `renderer.mesh.drawInstanced`'s
`instanceDataBuffer`, on slots no entity owns.
`VertexData` (in/out of `vertex()`): `position` (object-local, model-scale + morph
applied), `normal`, `uv`, `color`, `time`, and which vertex of which instance it
is: `instance_id` (`u32`, the instance's slot — `zero_object_data(v.instance_id,
lane)` reads the lanes the fragment gets as `shader_data`) and `vertex_index`
(`u32`, the vertex's index in its mesh). Both hold the same values in the colour
pass, the depth views, every shadow page and the previous-frame evaluation, so
geometry a hook builds from them — a blade placed and bent by its instance's
lanes, a strip shaped by its vertex's index — lands in the same place in each.
The rendering-instance-data guide has a worked blade field.
`ClipVertex` (in/out of `vertex_clip()`) — the post-transform vertex. Read context:
`clip_position`, `world_position`, `world_normal`, `uv`, `view_position`,
`view_direction`, `light_direction`, `light_color`, `light_intensity`,
`ambient_color`, `ambient_intensity`, `time`, `view_proj` (the view-projection of
the frame being evaluated), and where the vertex stands on its instance:
`local_position` (its position in the mesh, after `vertex()` and before the
instance's world transform), `object_origin` (the world position of the instance's
origin), `object_scale` (the world length of the instance's x, y and z axes) and
`object_axes` (`mat3x3<f32>`, the world direction of those axes, unit length) —
what a camera-facing card, a shard grown about its own base, or anything placed
per instance is built from. **Writable:** `clip_position` (snap /
jitter) and `color` (write a lit colour here for per-vertex Gouraud shading; the
engine forwards it to the interpolated `vertex_color` — left untouched, the mesh's
own vertex colour passes through).
`vertex()` and `vertex_clip()` run in the vertex stage, which reads the frame
uniforms (`uniforms.*`, and the `ClipVertex` members built from them), the scene
buffer (`zero_object_data`, `zero_object_visuals`, the node transforms), the
material interface and a `read` buffer property. The point, spot and area lights
(`zero_point_light`, `zero_punctual_projection`), the shadow maps, the
environment and light probes, the feature textures, the scene depth, colour and
reflection copies and a `@group(2)` buffer the shader declares `read_write` are
visible to the fragment stage alone: read them in `fragment()` or `surface()`. A
shader whose vertex hook reaches one fails to compile, and the error names the
stage, the binding and the calls that reach it.
### Each hook runs twice: this frame and the one before it
A vertex hook decides where the surface IS, so it also decides where the surface
WAS. The engine runs every hook a shader declares a second time, over the same
vertex standing one frame back — the previous-frame position stream, with
`time` = the clock that frame was drawn at and (for `vertex_clip()`)
`view_proj` = the view-projection it was drawn with. The motion vector the
renderer publishes is taken between the two results, so a displacement that
moves with time reports the velocity of the displaced surface, and a
displacement that does not move reports the motion of the transform alone
because the identical offset appears on both evaluations and cancels.
What that asks of a hook:
- **Take the frame from the argument.** `v.time` is the clock of the evaluation
being run; `uniforms.time` is always this frame's. A displacement built on
`uniforms.time` is identical on both evaluations, so its motion vector cancels
to exactly zero. The same holds for `v.view_proj` against
`uniforms.view_proj`: a clip hook that rebuilds its position from world space
through `uniforms.view_proj` reprojects the previous frame with this frame's
camera, and reports no camera motion at all. The wind helpers come in both
shapes for this reason — `zero_wind_sway_at_time(world_pos, stiffness, v.time)`
inside a hook, `zero_wind_sway(world_pos, stiffness)` outside one.
- **Anchor a wrap to `uniforms.time`.** A displacement that jumps discontinuously
— a falling drop that restarts at the top, a looping offset — differences its
jump as velocity on whichever frame it wraps. Compute the wrap COUNT from
`uniforms.time` and the position within the cycle from `v.time`, and both
evaluations share the same cycle, so the surface reports the speed it fell at.
`@builtin::shaders.rain_material` and `snow_material` do exactly that.
The per-vertex attributes (`world_normal`, `uv`, `color`) and the lighting and
camera-position context are this frame's on both evaluations, so a hook that
displaces along its normal reports the motion of the transform plus this frame's
displacement.
## Engine state — point lights & shadows
The common lighting state — the sun (`input.light_direction` / `light_color` /
`light_intensity`), the ambient (`input.ambient_color` / `ambient_intensity`), the
camera (`input.view_position` / `view_direction`), and `input.time` — is already on
`FragmentData`, so most shaders never touch the raw engine state. The one thing not
on `FragmentData` is the punctual-light array (its length varies), which is always
in scope. A scene can hold up to 1024 lights, so a fragment walks the ones the
light grid lists where it stands, which hold every light that can reach it:
| Expression | Type | Meaning |
|---|---|---|
| `zero_lights_near(world_pos)` | `ZeroLightsNear` | the lights that can reach a fragment at `world_pos`; `.count` of them |
| `zero_light_near(near, k)` | `u32` | the index of the `k`th of them, for `zero_point_light` |
| `uniforms.point_light_count` | `u32` | number of lights the frame shades under, every one of them |
| `zero_point_light(i)` | `PointLightData` | one point or spot light (fields below) |
| `zero_punctual_projection(i)` | `PunctualProjection` | a spot or area light's projection: `view_proj`, the frustum its cookie and shadow are placed by, and `cookie_layer` |
`PointLightData`: `position`, `radius`, `color`, `intensity`, `direction`,
`cone_outer_cos`, `cone_inner_cos`, `kind` (0 = point, 1 = spot).
Shadows come from `#include "@builtin::shaderModules.shadows"`:
- `compute_directional_shadow_factor(world_pos, normal, light_dir) -> f32` — 1.0 lit, 0.0 fully shadowed.
- `compute_spot_shadow_factor(...)` / `compute_point_shadow_factor(...)` / `compute_area_shadow_factor(...)` — one shadow shape each.
- `compute_punctual_shadow_factor(light_index, world_pos, normal) -> f32` — the row's own shape, read off its `kind`, for a loop that walks the punctual lights and does not want the switch.
- `zshadow_light_reach(light_index, world_pos, normal, object_channels) -> f32` — how much direct light that row puts on a surface on those light channels: the channel mask, falloff, cone or rect orientation, and the cosine, times intensity. Zero exactly where the light contributes nothing.
## Writing your own shading
The engine forces no lighting model — `zero_pbr_shade` is just one option. You can
light a surface yourself from the state above. Here is a complete custom-lit
(Lambert diffuse) surface with no PBR:
```wgsl
// @domain: surface
#include "@builtin::shaderModules.shadows" // exposes compute_directional_shadow_factor
fn fragment(input: FragmentData) -> vec4<f32> {
let n = normalize(input.world_normal);
let albedo = material.base_color.rgb;
// Directional (sun) light: Lambert N·L, attenuated by its shadow.
let sun = normalize(-input.light_direction); // surface -> sun
let sun_shadow = compute_directional_shadow_factor(input.world_position, n, input.light_direction);
var lit = input.light_color * input.light_intensity * max(dot(n, sun), 0.0) * sun_shadow;
// Point + spot lights: the ones that can reach this fragment.
let near = zero_lights_near(input.world_position);
for (var k = 0u; k < near.count; k = k + 1u) {
let pl = zero_point_light(zero_light_near(near, k));
let to_light = pl.position - input.world_position;
let dist = length(to_light);
let l = to_light / max(dist, 1e-4);
let atten = max(1.0 - dist / max(pl.radius, 1e-4), 0.0);
lit += pl.color * pl.intensity * max(dot(n, l), 0.0) * atten;
}
// Ambient fill, with the scene's ambient occlusion already in it.
lit += input.ambient_color * input.ambient_intensity;
return vec4<f32>(albedo * lit, material.base_color.a);
}
```
- **Unlit:** skip the lights and return your colour directly.
- **Toon / PS1 / custom:** start from the loop above and band, quantize, or warp
it however you like.
- **PBR:** build a `PbrSurface` and call `zero_pbr_shade` (see "Library includes")
instead of writing the BRDF by hand.
## Material properties — `properties.yaml`
A surface shader declares its material properties in a sibling `properties.yaml`.
The engine GENERATES the entire group(1) interface from it — the uniform struct
(`material.<name>`) plus a texture/sampler pair per declared texture. You write
no `@group`/`@binding` lines.
```yaml
properties:
- { name: base_color, type: color, default: [1, 1, 1, 1] }
- { name: roughness, type: float, default: 0.5, min: 0, max: 1 }
- { name: metallic, type: float, default: 0.0, min: 0, max: 1 }
- { name: emissive, type: color, default: [0, 0, 0, 1] }
- { name: base_color_texture, type: texture, default: white }
- { name: normal_texture, type: texture, default: flat_normal }
```
| `type` | WGSL field type | Notes |
|---|---|---|
| `float` | `f32` | optional `min`/`max` for editor sliders |
| `float2`/`float3`/`float4` | `vec2/3/4<f32>` | |
| `color` | `vec4<f32>` | linear RGB colour and alpha, default is `[r,g,b,a]`; the shader reads it as written |
| `int` | `i32` | |
| `bool` | `u32` | |
| `texture` | `texture_2d<f32>` + `sampler` | `default:` is a built-in texture name (`white`, `black`, `flat_normal`) |
A `color` holds linear RGB, the space lighting is computed in, the way `base_color` multiplies a colour texture's
linear sample. A colour picked from a picture or written as a hex code is sRGB, and written as it is it reads paler
and less saturated than picked: convert it first, `color.toLinear(assert(color.hex("#ff8800")))` (`{ r, g, b, a }`:
`#ff8800` is `{ r = 1, g = 0.246, b = 0, a = 1 }`), and write that.
In the shader you read scalars as `material.<name>` and sample textures as
`textureSample(<name>, <name>_sampler, uv)`. A texture slot with no bound texture
samples its default, so a shader can always sample — no presence flag needed.
`vertex()` reads the same material interface, so a height or deformation map
drives displacement from the slot it was bound to. A vertex stage has no implicit
derivatives, so sample it at an explicit mip there:
`textureSampleLevel(<name>, <name>_sampler, uv, 0.0)`.
## Sky shaders
A `// @domain: sky` shader draws the sky as a fullscreen pass. You write **only**
`fn fragment(in: SkyInput) -> vec4<f32>` — the colour along the view ray for each
pixel — and declare tunables in `properties.yaml` exactly like a surface shader
(read as `material.<name>`, textures via `textureSample(<name>, <name>_sampler, uv)`).
The engine generates the fullscreen `vs_main`/`fs_main`, the group(1) interface,
and reconstructs the world-space ray; there is no separate sky renderer — a sky
material runs through the same material pipeline, just on a fullscreen triangle.
A sky pipeline binds group(0) and the group(1) interface alone, so a sky that
reads a `buffer` property or any other `@group(2)` binding fails to compile, and
the error names the binding and the calls that reach it.
`SkyInput` (the only argument) carries:
| Field | Type | Meaning |
|---|---|---|
| `ray_dir` | `vec3<f32>` | normalized world-space ray through this pixel |
| `ndc` | `vec2<f32>` | screen position in NDC `[-1, 1]` |
| `sun_direction` | `vec3<f32>` | unit vector TOWARD the sun (from the directional light) |
| `camera_pos` | `vec3<f32>` | world-space camera position |
| `time` | `f32` | gameplay seconds: advances with the world at `engine.timeScale` and stands still while gameplay is paused (edit mode starts paused), so an effect slows and stops with the world |
| `constants` | `SkyConstants` | what the shader's `sky_constants()` returned for this draw (zeros when it declares none) |
```wgsl
// @domain: sky
fn fragment(in: SkyInput) -> vec4<f32> {
let up = max(in.ray_dir.y, 0.0);
let col = mix(material.horizon_color.rgb, material.zenith_color.rgb, up);
return vec4<f32>(col * material.exposure, 1.0);
}
```
Day/night comes from `sun_direction.y` (the directional light's elevation), so a
sky shader never needs a time-of-day uniform — orient the light and the sky
follows. The builtin sky shaders (`procedural_sky`, `solid_sky`, `cubemap_sky`,
`equirect_sky`) are all authored this way; author a custom sky material and point
a `Skybox` component at it.
`fragment()` runs once for every pixel of the view and every texel of each
environment-capture face. A value that holds for the whole draw (a colour mixed
from the material and the sun's elevation, a `cos` of an angular size, an
exponent derived from a property) goes in an optional
`fn sky_constants(in: SkyConstantsInput) -> SkyConstants`, which the generated
vertex stage evaluates once per draw. `fragment()` reads its result as
`in.constants`. `SkyConstantsInput` carries `sun_direction`, `camera_pos` and
`time` (the same values `SkyInput` holds), `sky_constants()` reads the material
as `material.<name>` like `fragment()`, and `SkyConstants` is eight `vec4<f32>`
slots, `v0` to `v7`, packed however the sky likes:
```wgsl
// @domain: sky
fn sky_constants(in: SkyConstantsInput) -> SkyConstants {
let day = clamp(in.sun_direction.y * 2.0, 0.0, 1.0);
var c: SkyConstants;
c.v0 = vec4<f32>(mix(vec3<f32>(0.02), material.zenith_color.rgb, day), day);
c.v1 = vec4<f32>(mix(vec3<f32>(0.01), material.horizon_color.rgb, day), 0.0);
return c;
}
fn fragment(in: SkyInput) -> vec4<f32> {
let up = max(in.ray_dir.y, 0.0);
return vec4<f32>(mix(in.constants.v1.rgb, in.constants.v0.rgb, up), 1.0);
}
```
A branch on an `in.constants` value is taken or skipped by every pixel of the
draw together, so a term that only contributes some of the time (stars at
night) costs nothing while it is off. The vertex stage has no implicit
derivatives, so a texture read in `sky_constants()` is `textureSampleLevel`.
## Post-process & screen shaders
`// @domain: post-process` and `// @domain: screen` shaders are authored exactly
like surface and sky — you write **only** `fn fragment(in: PostInput) -> vec4<f32>`
(post) or `fn fragment(in: ScreenInput) -> vec4<f32>` (screen) and declare tunables
in `properties.yaml` (read as `material.<name>`, textures via
`textureSample(<name>, <name>_sampler, uv)`). The engine generates the fullscreen
`vs_main`/`fs_main` and the group(1) material interface. **There are no
`uniforms.params[i]` slots** — that convention is retired; every input is a named
property.
A **post-process** shader's `properties.yaml` carries one further, top-level key
beside `properties:` — `space`, either `light` (the default) or `display`:
```yaml
space: display
properties:
- { name: amount, type: float, default: 1.0, min: 0, max: 1 }
```
`light` is the scene's linear radiance — what every surface, the sky and every
`afterLighting` feature write, unbounded above 1.0. `display` is the code values
the `__display__` pass leaves behind after applying exposure and its tone curve
— the 0-1 range a colourist's tool and a dithered or quantized effect expect.
`postprocess.add`'s `opts.space` overrides what the asset declares for one
registration. The chain converts at both ends of an effect that runs on a
segment in the other space, so a light-space vignette still darkens a frame
graded after `__display__`, and a lookup table built from a `.cube` export
always sees the code values it was authored against, wherever the chain
runs it.
`space` also decides which layer a registration that names none lands on: a
`light` shader grades `scene`, the segment holding the scene's own light, and a
`display` shader grades `display`, the picture after `__display__` and before
the UI lands on it. So `postprocess.add("glow", asset.resolve("~.shaders.bloom",
"shader"))` thresholds highlights that still stand above 1.0, and a lookup
table registered the same way reads the code values it was built for and leaves
every screen in the colours it declares. `opts.layer` names one explicitly —
`"all"`, after the UI, is what an effect that must grade the interface says.
A post-process shader may also declare the post stage of the frame it runs in,
`stage:` — `optics` (the `scene` layer), `display` (the `display` layer) or
`ui` (the `all` layer); `renderer.post.stages()` lists every stage with the
colour space and resolution its picture is held in. A registration takes the
declared stage unless `opts.stage` or `opts.layer` names another seam, and a
stage the frame does not have, or one only render features run in, is refused
naming the stages there are:
```yaml
space: display
stage: display
properties:
- { name: amount, type: float, default: 1.0, min: 0, max: 1 }
```
A **post-process** shader's `properties.yaml` carries a further, optional
top-level key — `passes:`, the pass list the frame draws through, in order:
```yaml
space: light
passes:
- { name: bright, entry: bright, scale: 0.5, inputs: [scene] }
- { name: down1, entry: downsample, scale: 0.25, inputs: [bright] }
- { name: up0, entry: upsample, scale: 0.5, inputs: [down1, bright] }
- { name: composite, entry: composite, inputs: [scene, up0] }
```
Every pass's entry — `fn <entry>(in: PostInput) -> vec4<f32>` — lives in the
one `shader.wgsl` the assetType recompiles on. `scale` is the fraction of the
view a pass draws at; the LAST pass takes none, since it draws at the view's
size and writes the chain's output. `inputs` names what a pass reads —
`"scene"`, the chain's input for this effect, or the name of an EARLIER pass —
and `inputs[0]` is what `zero_sample_scene(uv)` reads inside that pass's body;
every input, `inputs[0]` included, is also reachable by index through
`zero_pass_input(i, uv)` and `zero_pass_input_size(i)`, below. The one place
the two reads of `inputs[0]` disagree: in an effect whose declared space
differs from the segment it draws on, `zero_sample_scene` converts a `scene`
input into that space while `zero_pass_input(0, uv)` hands back the segment's
own, unconverted value. A shader declaring no `passes:` is one pass whose
entry is `fragment`, and its generated WGSL is byte-identical to what it is
without the key.
A pass that is not the last writes a scratch target the effect owns — in the
space the shader declared, view-sized times its own `scale` — allocated on the
first draw and replaced whenever the size changes. `engine.params`, below,
carries the drawing pass's own index, its scale, and its first input's texel
size, so one shader body serves every level of a chain built this way.
The worked example is the bloom: `@builtin::shaders.post.bloom` declares
twelve passes — a threshold pass, five downsamples, five tent upsamples and a
composite — out of four entries, since the five downsamples share one and the
five upsamples share another, each level reading the one below it through
`zero_pass_input` and telling itself apart by `engine.params`. Its own README
walks the declaration.
A declaration is refused at compile time — reported through `compileStatus()`,
`postprocess.add`'s second return value, or a live effect's
`status()[i].error` — for: an entry `passes:` names that `shader.wgsl` does
not define; an input naming a later pass, or naming neither `scene` nor an
earlier pass; a `scale` outside `(0, 1]`; a `scale` declared on the last pass;
a `buffer` property (group(2) carries the pass inputs); or a shader whose
group(0) plus declared texture slots plus widest pass's inputs ask for more
than the 16 FRAGMENT sampled textures one shader stage has.
A **post-process** shader runs as a full-screen pass over the already-rendered
frame. Its `fragment` argument and framework helpers:
`PostInput`:
| Field | Type | Meaning |
|---|---|---|
| `frag_coord` | `vec4<f32>` | framebuffer pixel coords (`@builtin(position)`) |
| `uv` | `vec2<f32>` | screen UV, `(0,0)` top-left … `(1,1)` bottom-right |
Framework helpers (always in scope — no include needed):
| Call | Returns | Meaning |
|---|---|---|
| `zero_sample_scene(uv)` | `vec4<f32>` | the rendered scene colour at `uv`, converted into the space this effect declared (`zero_scene_light` is the same call, for content that reads it by that name) |
| `zero_scene_depth(uv: vec2<f32>)` | `f32` | scene depth at `uv` — a screen UV here, where the surface domain's `zero::scene_depth` module takes a pixel coordinate under the same name (exact texel — an R32Float copy). Reversed: 1.0 is the near plane, 0.0 the far plane and empty sky, so a nearer surface reads GREATER and `d > 0.0` means "a surface was drawn here" |
| `zero_linear_depth_through(d, near, far, engine.view_proj)` | `f32` | that sample as a distance in metres along the camera's view axis, under a perspective or an orthographic camera — the one inversion of the reversed range (`zero_linear_depth(d, near, far)` is its perspective case alone) |
| `zero_scene_motion(uv)` | `vec2<f32>` | per-pixel screen-space motion vector at `uv` |
| `zero_pass_input(i, uv)` | `vec4<f32>` | the pass's declared input `i`, at `uv`. Input 0 is what `zero_sample_scene` reads |
| `zero_pass_input_size(i)` | `vec2<f32>` | that input's size in texels, so one entry serves every level of a chain |
| `zero_display_exposure()` | `f32` | the exposure in force: the gain `__display__` multiplies the scene's light by before its curve, the value this frame's display pass draws with (1 where the chain carries none). Light times it is light in the units the picture shows it at, so a threshold judged against `light * zero_display_exposure()` picks the same highlights at every exposure |
Engine data uniform `engine` (a `PostEngineData`):
| Field | Type | Meaning |
|---|---|---|
| `resolution` | `vec4<f32>` | `x`=width, `y`=height, `z`=1/width, `w`=1/height |
| `time` | `vec4<f32>` | `x`=elapsed seconds, `y`=dt, `z`=frame index |
| `view_proj` | `mat4x4<f32>` | this frame's view-projection |
| `prev_view_proj` | `mat4x4<f32>` | last frame's view-projection (motion) |
| `inv_view_proj` | `mat4x4<f32>` | inverse view-projection (world-space reconstruction) |
| `params` | `vec4<f32>` | `x`=this pass's own index, `y`=the scale it draws at, `zw`=1/size of its first input — for a pass the post-process chain draws. A pass a `renderFeature` fragment pass draws carries that feature's `spec.params` here instead; a shader declaring `passes:` is always drawn by the chain, so the two never meet in one shader. |
| `exposure` | `vec4<f32>` | `x`=the exposure in force (what `zero_display_exposure()` reads) |
The engine fills all of it **per render target**. A registered effect runs over
the live viewport's frame AND over every offscreen one — a `capture` from a
world-space station or orbiting an entity, a capture of a named camera, a
render-to-texture camera — and in each of those, `view_proj`, `prev_view_proj`
and `inv_view_proj` are the camera THAT render was drawn from, while
`resolution` is that target's own pixel size. So a pass reconstructing world
space from `zero_scene_depth(uv)` reconstructs against the station and lens the
capture asked for, and an offscreen frame is a picture of the effect as it
stands at that station. (A render feature's passes reach an offscreen render the
same way, against the same camera — the `renderFeature` assetType README carries
that half, including `@frame.camera`, its own per-target camera input.)
An offscreen render keeps no view history of its own, so `prev_view_proj` there
holds that same matrix rather than the frame before it, and a pass taking camera
motion from the two reads none. On the viewport the pair is a frame apart, so
camera motion is the one reading a capture answers differently from the screen.
`postProcessing = false` on a capture is the one control that takes the chain
off the frame it returns; it names the camera property the ephemeral capture
camera is built with, so it scopes to that one capture. What a capture opts out
of is the registered effects; `__display__` still runs on it, so the exposure
and tone curve that turn the scene's light into a picture are on what a capture
returns.
A diagnostic capture (`albedo`, `normal`, `motion_vectors`, …) bypasses
`__display__` itself, since a channel like that is a reading of the scene
rather than a graded picture of it. The environment-cube capture bypasses it
too, and for the same reason: a probe face carries radiance into the cube
array, and `__display__` would fit that radiance to a display's code values —
a picture of it rather than the value the PBR indirect term multiplies into
its lighting.
A complete post-process effect — a vignette via `fragment()` + `properties.yaml`:
`properties.yaml`:
```yaml
properties:
- { name: intensity, type: float, default: 0.5, min: 0, max: 1 }
- { name: radius, type: float, default: 0.75, min: 0, max: 1 }
- { name: softness, type: float, default: 0.45, min: 0, max: 1 }
```
`shader.wgsl`:
```wgsl
// @domain: post-process
fn fragment(in: PostInput) -> vec4<f32> {
let color = zero_sample_scene(in.uv);
let intensity = max(material.intensity, 0.01);
let radius = max(material.radius, 0.01);
let softness = max(material.softness, 0.01);
let center = in.uv - vec2<f32>(0.5);
let dist = length(center);
let vignette = smoothstep(radius, radius - softness, dist);
let darkened = color.rgb * mix(vec3<f32>(1.0), vec3<f32>(vignette), intensity);
return vec4<f32>(darkened, color.a);
}
```
The asset above **defines** the effect. It enters the frame when something
registers it:
```luau
postprocess.add("vignette", asset.resolve("~.shaders.vignette", "shader"), { priority = 200 })
```
The registration names no layer, so the effect grades the layer its `space`
belongs to: this vignette reads light, so it runs on `scene`, over the picture
before the UI reaches it. Registering it is the one step that makes it render,
so a world's chain holds exactly the passes its content asked for. Editing the shader body afterwards
recompiles every effect registered from it, in place — the pass keeps its enabled
state, priority, layer, tuned property values and bound textures, and draws the
new pixels on the next frame. `postprocess.remove(name)` takes it back out of the
chain and leaves the asset alone.
Drive the named properties at runtime with `postprocess.setProperty(name, prop,
value)` and bind texture properties with `postprocess.setTexture(name, prop, path)`
(both keyed by the **property name**, never a slot index — see the post-process
section of `man shaders` / the `postprocess.*` API).
Read them back with `postprocess.describe(name)`: it carries the chain state
`postprocess.status()` lists for that effect, the `properties` schema it was
registered with, and the `values` each of those properties currently holds. Before
the effect exists, `shaderRef:getProperties()` on the asset names the same schema,
so what a pass takes is answerable either side of the registration.
A **screen** shader is a standalone full-screen visual (UI background, procedural
pattern). `ScreenInput` mirrors `PostInput` (`frag_coord` + `uv`); there is no scene
to sample, so it has no `zero_sample_scene` helpers. Its `engine` uniform is a
`ScreenEngineData`:
| Field | Type | Meaning |
|---|---|---|
| `time` | `f32` | gameplay seconds: advances with the world at `engine.timeScale` and stands still while gameplay is paused (edit mode starts paused), so an effect slows and stops with the world |
| `scroll` | `f32` | UI scroll offset |
| `resolution` | `vec2<f32>` | render-target size in pixels |
| `mouse` | `vec2<f32>` | mouse position |
| `velocity` | `f32` | scroll velocity |
```wgsl
// @domain: screen
fn fragment(in: ScreenInput) -> vec4<f32> {
let t = engine.time * 0.5;
let col = vec3<f32>(
sin(in.uv.x * 3.14159 + t) * 0.5 + 0.5,
sin(in.uv.y * 3.14159 + t * 0.7) * 0.5 + 0.5,
material.brightness
);
return vec4<f32>(col, 1.0);
}
```
## Library includes
A library module is a **`.shaderModule` asset**: it only EXPOSES functions and
structs and does nothing on its own (exactly like a Unity `.cginc`). You must
call what you include. Include one by its identity, the same string you would
hand `asset.resolve`:
| Include | Exposes |
|---|---|
| `@builtin::shaderModules.pbr_shading` | `struct PbrSurface { albedo, normal (world), metallic, perceptual_roughness, emissive, occlusion }`; `zero_pbr_default_surface(FragmentData) -> PbrSurface`; `zero_pbr_shade(FragmentData, PbrSurface) -> vec3<f32>` (the full forward PBR lighting); `zero_pbr_indirect_visibility(FragmentData, PbrSurface) -> ZpbrIndirectVisibility` (the share of the surroundings' light each indirect lobe takes from the scene's ambient occlusion: `diffuse` with the light a crease bounces back, `specular` from the open cone around the bent normal); `zero_pbr_shade_with_visibility(FragmentData, PbrSurface, ZpbrIndirectVisibility) -> vec3<f32>` (the same lighting with that share stated by the caller) |
| `@builtin::shaderModules.shadows` | `compute_directional_shadow_factor(...)`, `compute_spot_shadow_factor(...)`, `compute_point_shadow_factor(...)`, `compute_area_shadow_factor(...)`, `compute_punctual_shadow_factor(...)`, `zshadow_light_reach(...)`; each light's `_with_kernel` form takes a shared `ZShadowKernel` (`zshadow_kernel()`) so several lookups of one fragment derive its filter rotation once |
| `@builtin::shaderModules.engine_bindings` | the group(0) declarations (camera, transforms, lights, shadow maps) — surface shaders get this implicitly; include it directly only for advanced raw access |
| `@builtin::shaderModules.shadow_bias` | `zshadow_offset_receiver(...)`, `zshadow_directional_depth_bias(...)` and the rest of the bias math, as pure functions |
| `@builtin::shaderModules.texturing` | `ztex_stochastic(...)` (tile without a visible repeat grid), `ztex_blend_normals(...)` (reoriented normal blending) |
| `@builtin::shaderModules.depth` | `zero_linear_depth_through(depth, near, far, view_proj)` — the one reading of a stored depth, for the projection the view-projection states; `zero_linear_depth` and `zero_linear_depth_orthographic` are its two cases |
| `@builtin::shaderModules.ambient_light` (`zero::ambient_light`) | `zpbr_surroundings_irradiance(world_pos, n, flat_ambient)` (the diffuse light the surroundings hand a surface with normal `n`: the flat ambient, the captured sky and reflection probes, and the baked light-probe volumes, as the PBR model reads it) and `zpbr_surroundings_in_scatter(world_pos, flat_ambient)` (that light averaged over every direction: what a participating medium scatters). A compute pass binds `environment_map`, `environment_sampler` and `zero_environment` to `@scene.environment`, `@scene.environmentSampler` and `@scene.reflectionProbes` |
| `@builtin::shaderModules.gbuffer_material` (`zero::gbuffer_material`) | `struct ZeroGbufferMaterial { base_color, metallic, perceptual_roughness }`; `zero_gbuffer_unpack_material(texel: vec4<u32>) -> ZeroGbufferMaterial`; `zero_gbuffer_material(t: texture_2d<u32>, coord: vec2<i32>) -> ZeroGbufferMaterial`, `zero_gbuffer_base_color`/`zero_gbuffer_metallic`/`zero_gbuffer_roughness(t, coord)` — the one unpacking of `@scene.material`'s `Rgba8Uint` texel into what a surface reflects |
| `zero::scene_depth` | `zero_scene_depth(px: vec2<f32>)`, `zero_scene_depth_uv(uv: vec2<f32>)`, `zero_scene_depth_texel(px: vec2<i32>)`, `zero_scene_depth_size()`, `zero_scene_depth_is_sky(d: f32)`, `zero_scene_world_position(px: vec2<f32>)`, `zero_scene_world_normal(px: vec2<f32>)`, `zero_scene_depth_fade(px: vec2<f32>, world_position: vec3<f32>, fade_distance: f32)` — every `px` is a pixel coordinate in two components, so a surface fragment passes `input.frag_coord.xy`, not the `vec4<f32>` `FragmentData.frag_coord` holds. The depth already drawn UNDER a surface fragment, and the world position and geometric normal of the surface standing there. What a decal projector box, a soft particle, a forcefield or water intersection foam shades against. Surface domain; the depth belongs to the view being drawn |
| `zero::scene_color` | `zero_scene_color(px: vec2<f32>)`, `zero_scene_color_uv(uv: vec2<f32>)`, `zero_scene_color_texel(px: vec2<i32>)`, `zero_scene_color_size()`, `zero_scene_color_refract(px: vec2<f32>, offset_px: vec2<f32>)`, `zero_scene_color_refract_uv(px: vec2<f32>, offset_fraction: vec2<f32>)` — every `px` is a pixel coordinate in two components, so a surface fragment passes `input.frag_coord.xy`, not the `vec4<f32>` `FragmentData.frag_coord` holds. The scene colour already composed BEHIND a surface fragment, and the displaced read that bends it. What heat haze, a refraction shockwave, a cloaking or shield warp, glass and ice shade with. Surface domain; the colour belongs to the view being drawn, and is the lit opaque scene as it stood immediately before the transparent pass — so a `blend`-queue surface reads this frame's, an opaque one reads the previous frame's, and two overlapping blended surfaces both read the same picture and do not bend each other |
| `zero::projection` | `zero_projection_is_orthographic(view_proj: mat4x4<f32>) -> bool`, `zero_projection_parallel_ray(view_proj: mat4x4<f32>) -> vec3<f32>`, `zero_projection_view_direction(view_proj: mat4x4<f32>, camera_position: vec3<f32>, world_position: vec3<f32>) -> vec3<f32>`: what a view-projection states about how it projects: whether it is orthographic, the direction an orthographic one sees along (pointing back toward the camera), and the direction a camera sees a world point from under either projection, the value `FragmentData.view_direction` holds. Pure math with no bindings, so a surface, post-process or compute shader passes the view-projection it holds (`uniforms.view_proj`, `engine.view_proj`, or a uniform of its own) |
`asset.list("shaderModule")` enumerates the `.shaderModule` assets in this
engine, builtin and world-authored alike, and each one's README says what it
exposes. Those are the `@builtin::shaderModules.*` rows above; a module you
write is included exactly the same way, by its own identity or by the
`~.name` form.
The `zero::`-named rows come from the renderer itself. Their WGSL is baked into
the engine binary and registered under those names before any asset exists, so
they answer an `#include` in every engine and appear in no asset index. A shader
that includes one records no dependency on it: the engine reading the shader is
the engine that already holds the module.
Two spellings reach the same `.shaderModule`: a module's declared alias
answers as well as its identity, so a shader written against `zero::pbr`
compiles and pins the asset that name resolves to. Prefer the identity: it
says which asset travels with the content.
Whichever punctuation an include is written with — `#include "name"`,
`#include 'name'` or `#include <name>` — names the same module and records the
same dependency.
The engine's own scaffolding — `zero::surface`, `zero::sky`, `zero::post`,
`zero::screen` — is injected by the compiler around the matching domain and is
never written by hand. The surface framework puts `safe_normalize`,
`mat3_from_mat4` and `apply_normal_map(FragmentData, tangent_normal)` in scope
for every surface shader.
## Shader features — compile-time variants
A **property** is runtime uniform data (a colour, a float, a texture): set per
material, read as `material.<name>`, no recompile. A **feature** decides which
code is in the program at all. Use a property when the value changes at runtime,
a feature when it changes what the shader is.
Declare features in `properties.yaml`, alongside the properties:
```yaml
properties:
- { name: base_color, type: color, default: [1, 1, 1, 1] }
features:
- { name: PARALLAX, default: false, description: silhouette parallax march }
- { name: DETAIL, default: true, description: detail-normal overlay }
```
Guard the code each feature owns with `#ifdef`:
```wgsl
fn surface(input: FragmentData) -> PbrSurface {
var s = zero_pbr_default_surface(input);
#ifdef PARALLAX
s.normal = parallax_normal(input); // absent when PARALLAX is off
#else
s.normal = input.world_normal;
#endif
return s;
}
```
`#ifdef` / `#ifndef` / `#else` / `#endif`, one per line, and they nest. The
directives resolve before the WGSL reaches the compiler, so the program holds
exactly the lines its feature set selected and every other line is an empty one,
which keeps a compiler message on the line number the file shows. An unbalanced
block is a compile error naming its line.
A material is what chooses a feature set, so `features:` shapes the shaders a
material draws with — the surface, sky and screen-space domains.
A material names the features it wants in its `mat.yaml`:
```yaml
name: worn_rock
shader: "@builtin::shaders.pbr_layered"
features:
PARALLAX: true # on, over the shader's default
DETAIL: false # off, against the shader's default
```
or through `asset.create`:
```lua
asset.create("material", "worn_rock", {
shader = "@builtin::shaders.pbr_layered",
features = { PARALLAX = true, DETAIL = false },
})
```
### A property that switches a feature on
A feature can also be switched on by a material's property values. List the
properties under the feature's `properties:`, and the feature is on for every
material whose value for one of them stands away from that property's declared
default:
```yaml
properties:
- { name: clearcoat, type: range, default: 0.0, min: 0, max: 1 }
features:
- { name: CLEARCOAT, properties: [clearcoat], description: the coat lobe }
```
```wgsl
#ifdef CLEARCOAT
s.clearcoat = material.clearcoat;
#endif
```
A material at `clearcoat: 0` draws a program holding none of the coat's code; one
that sets `clearcoat: 0.6`, in its `mat.yaml`, through `asset.create` or with
`renderer.material.setProperty`, draws the `CLEARCOAT` variant, and setting it
back to 0 returns it to the base program. The property stays a runtime value the
program reads; the feature decides only whether the code reading it is there.
This is how a shader with costly optional parts keeps every material that leaves
them at rest on the cheap program without each material having to name a
feature. A material naming the feature off under `features:` keeps it off
whatever the property holds. Write the guarded code so that, with the feature
off, the shader computes what it would with the property at its default: the
material sees one look either way, and pays for the part only where it shows.
Each distinct set of switched-on features is a program to compile the first time
a material reaches it. A material keeps the programs its values switched it onto
compiled while it lives, so a property crossing back and forth reuses them; when
the shader's variant budget is spent, the ones no material draws with are given
back first. A `properties:` entry is a value property (`float`, `range`, `int`,
`bool`, `color`, `float2` to `float4`) the shader declares; one naming no declared
property, or a texture or buffer, is refused by name.
### A texture that switches a feature
A feature can be switched by the textures a material binds. `textures:` lists
texture slots, and the feature is on for every material binding a texture in one
of them other than the slot's declared default; `shared_texture:` lists two or
more slots, and the feature is on for every material binding one and the same
texture in all of them. This is how a shader samples only the maps a material
binds:
```yaml
properties:
- { name: base_color_texture, type: texture, default: white }
- { name: rm_texture, type: texture, default: white }
- { name: ao_texture, type: texture, default: white }
features:
- { name: BASE_MAP, default: true, textures: [base_color_texture], description: the base colour map }
- { name: AO_MAP, default: true, textures: [ao_texture], description: the occlusion map }
- { name: PACKED_AO, shared_texture: [rm_texture, ao_texture], description: occlusion read from the rm map }
```
```wgsl
var base_texel = vec4<f32>(1.0); // what the white default stands for
#ifdef BASE_MAP
base_texel = textureSample(base_color_texture, base_color_texture_sampler, uv);
#endif
```
A switched feature, by value or by texture, is decided by what the material
holds: on where one of its switches holds, off where none does, and a material
naming the feature under `features:` decides it over the switch. A map feature
declared `default: true` keeps the shader's own default configuration sampling
every map, so a material whose feature set the variant budget has no room for
still draws its maps. A `textures:` or `shared_texture:` entry naming no
declared texture slot, and a `shared_texture:` listing one slot, are refused by
name.
The engine compiles one program per feature set its materials ask for, named
`<shader>#<features>`. Materials sharing a set share that program. A feature the
shader does not declare selects nothing, so a stray name cannot make the engine
compile anything, and the compile names both halves of a mismatch: a guard on a
name `features:` does not declare, and a declared name no guard tests.
**Permutations are the cost.** A shader with n features can express 2^n
programs, each its own compile and pipeline. One shader holds at most 32
variants; a material asking for a set past that draws with the shader's default
features and the engine logs which shader ran out. Read what a scene is
spending:
```lua
for _, s in renderer.shaderVariants() do
print(s.shader, #s.variants, "of", s.budget)
print(renderer.variantSource(s.base.program)) -- what it ships as
for _, v in s.variants do
print(v.program, renderer.variantSource(v.program)) -- what one asked for
end
end
```
A shader registers under every key it answers to — its identity, its guid, each
alias — and each key is a row. A variant lands on the row for the key the
material named, which for authored content is the guid, so take a program name
from the report rather than composing one.
## Materials reference shaders
A `.material` applies a shader to entities. Its `mat.yaml` references the shader by
its identity string and supplies property values:
```yaml
name: my_material
shader: "@builtin::shaders.pbr"
render:
blend: alphaBlend
cull: back
depth_write: false
properties:
roughness: 0.5
base_color: [1.0, 1.0, 1.0, 1.0]
textures:
base_color_texture: { ref: "default:white" }
```
`shader:` is a plain identity string, and a colour is an array. A key's
value is the rest of its line, so every value above stands alone on one.
### The render block
Render behaviour (the blend equation, alpha
cutout, which faces are drawn) lives in the `render:` block and drives the
pipeline, not the fragment. These are the keys that reach the pipeline from both
authoring routes, and the values each one takes:
| Key | Values | Sets |
|-----|--------|------|
| `blend` | `opaque` `transparent` `alpha` `alphaBlend` `additive` `add` `cutout` `alphaCutout` `premultiplied` `premultipliedAlpha` `multiply` | the equation the fragment composites with |
| `cull` | `back` `front` `none` `off` `disabled` | which faces the rasterizer keeps |
| `depth_write` | `true` `false` | whether the draw writes depth |
| `depth_compare` | `less` `lessEqual` `lequal` `equal` `greater` `greaterEqual` `gequal` `always` `never` | the test a fragment's depth passes |
| `topology` | `triangleList` `lineList` `lines` `line` | what consecutive vertices assemble into |
| `queue` | a whole number: `2000` opaque, `3000` transparent, `4000` overlay | the draw order of the material; a number inside a queue orders within it |
`blend` is also spelled `type`, and the two name one key. Its equations, over
the fragment the shader returned (`src`) and the value already in the target
(`dst`): `opaque` is `src`; `transparent`, `alpha` and `alphaBlend` are
`src * src.a + dst * (1 - src.a)`; `additive` and `add` are `src + dst`, the
mode a glow, a light shaft or a spark card is written for; `premultiplied` and
`premultipliedAlpha` are `src + dst * (1 - src.a)`, taking a source that already
carries its own coverage; `multiply` is `src * dst`, which is how a soot, grime
or shadow card darkens what stands behind it. `cutout` and `alphaCutout` keep
the fragment or discard it on its alpha and go on writing depth.
The same block is what `asset.create`'s `render` argument takes, key for key and
value for value, so one material named in Luau and the same material standing on
disk ask for their pipeline with the same words. `depth_write` is the one key
whose value takes the shape of its route: the word `false` on a `mat.yaml` line,
the boolean `false` in the Luau table.
```lua
asset.create("material", "heat_plume", {
shader = "@builtin::shaders.pbr",
render = { blend = "alphaBlend", cull = "none", depth_write = false },
})
```
Where a value outside a key's set is reported is what the two routes differ on.
`asset.create` refuses the call on any of these keys and names the set it
measured the value against, or for `queue`, that it takes a number. A
`mat.yaml` standing on disk registers either way, each key falling back to a
value of its own: `blend` to `opaque`, `cull` to `back`, `depth_write` to
`true`, `depth_compare` to `less`, `topology` to `triangleList` and `queue` to
`2000`. The engine log carries the blend fallback, naming the value and
the set it was measured against, so a `mat.yaml` value is worth putting through
`asset.create` once — that is where every key answers with its set.
## Registration & hot reload
A `.shader` is compiled by the asset system, not by the renderer: writing
`shader.wgsl` or `properties.yaml` fires the `.shader` assetType's `onChange`,
which compiles it: the body + the property schema are read, the group(1)
interface and (for surface) every render-mode entry point are generated,
`#include`s are expanded, the result is **validated**, and the shader is
registered. Validation
happens at this compile step — the whole point of "compile". A shader with invalid
WGSL (an unknown field, a type mismatch, an undefined function) **fails to compile**
with the real reason — readable directly from Luau via
`shaderRef:compileStatus()` (`{ status = "compiled" | "failed" | "pending",
error?, pending? }`), and also in the author log and `asset.inspect`; it is **not**
registered. This fires on create and on every edit, so saving a `.wgsl` recompiles
with no world reload; the change is visible on the next frame. Compilation is
async, so a `compileStatus()` of `"pending"` immediatel