# agentSkill A skill packages what an agent needs to do one job well: the instructions, the assets and toolboxes the job runs through, and — for jobs with distinct branches — subskills that go deeper. Skills advertise themselves. Every registered skill's name and one-line description rides the agent's tool responses, so an agent discovers what this world knows how to do without being told to look. Invoking one returns the instructions plus the exact identities and tool names the job needs. ## Shape ``` <name>.agentSkill/ skill.yaml # the manifest: description, when, checks, dependencies, order instructions.md # the prose an agent receives on invoke README.md # what this skill is, for a human reading the tree .metadata # tags <sub>.agentSkill/ # optional subskills, reachable as <name>/<sub> ``` ## The manifest ```yaml $schema: zero/agent-skill/v1 description: Making a character a person can actually play. when: > Reach for this when the player should look like something else, or when you are about to build a camera and a body and wire them together by hand. checks: - check: Holding a movement key in play moves the body across the ground. observe: Read its position before and after holding the key; the numbers differ. - check: The body animates while it moves. observe: A time-sampled collage from a camera outside the player, while the key is held. dependencies: - "@builtin::tools.scene" - "@builtin::tools.scene.player" - "@builtin::components.PlayerPrototype" - "@builtin::docs.core.multiplayer" order: [spawn-points, camera-rig] ``` `description` is the line the roster carries, and the whole basis on which a skill is chosen. Name the job the skill makes a reader competent at, in the outcome they are after — "getting a scene's player and view right", "making content the engine will load". A reader scanning the roster is matching it against the job in front of them, so the closer it reads to their own sentence about that job, the sooner they open it. `when` is the longer trigger — the situations that should send a reader here, in the words they would use for their own task. `checks` is what the reader is answerable for at the end, and it leads the rendered skill, ahead of the procedure. Each entry pairs a **condition that must hold** with **where it is seen**, and both halves are required — a condition alone is satisfied by any glance, and the wrong vantage is how a job passes while every one of its faults sits outside the frame. Write them as observations someone else could repeat and disagree with: a number that changed, a thing visible in a named shot, a count that came back as one. A skill publishes only once it carries at least one. `stages` models a skill whose passes run in order, when the order is load-bearing. Each stage is `{ name, detail?, checks? }`, and a check that sits in a stage is answered after that pass has run — which is the difference between a criterion that means something and one that misleads. A skill with no meaningful order leaves `stages` out and lists its checks flat; a skill that has one may do both, and the flat checks then read as the final gate. `verdicts` names the fixed set of outcomes a reader reports — at the end of each stage when the skill has stages, and once at the end when it does not. Each is `{ verdict, means }`: the word, and what reporting it asserts. Both halves are required for the reason both halves of a check are. A verdict is a check one level up, deciding a whole pass rather than one condition, and a bare `refine` is a label each reader fills in privately — so two runs report the same word about different states and nothing downstream can tell them apart. Saying what it means settles the boundary in the skill instead of in whoever happens to be reading it. A check marked `perRun: true` is one whose content is written for the job at hand: the manifest can say *keep a list of every feature this object must have and answer against it*, but not what is on that list, because that is per run. It renders with that instruction attached. ```yaml verdicts: - verdict: continue means: Every check in the stage held, and the next stage can build on it. - verdict: refine means: A check failed on something this stage can still fix. Run it again. - verdict: stop means: A check failed on something an earlier stage decided. Go back to it. stages: - name: blockout detail: Mass and proportion only. checks: - check: Every part on the identity list has a stand-in at roughly the right size. observe: A capture beside the reference, at the same angle. perRun: true - name: material checks: - check: Metal reads as metal and cloth as cloth. observe: The metallic and roughness passes, after the lighting stage has run. ``` `dependencies` is one flat list of asset **identities**: a toolbox, a single tool, a guide, a component, a module, another skill. Each entry is resolved when the skill is opened and reported with what it turned out to be, so a skill never declares "this one is a toolbox" — the asset says so, and the reader groups by that. Identities are what make a skill portable. They pin like any other reference, so installing a skill into another world brings the toolboxes and guides it runs through along with it. A bare name (`scene`) or a guide path (`core/multiplayer`) resolves to nothing and is refused before publish. A toolbox entry lists the tools it carries, and a tool entry carries its own signature, so the instructions never restate a call contract that can drift. An identity this world lacks is reported as missing rather than dropped. Write instructions that name the toolboxes and tools a job runs through, and leave the call form to the reader — an agent may be holding the MCP tool surface, the shell, or Luau, and the signatures are listed for it either way. ## Subskills A `<sub>.agentSkill/` folder nested inside a skill is a subskill. It has the same shape as its parent and is invoked as `parent/sub`. Subskills are reachable only through their parent: they are absent from the roster and from `skills.list`, and appear in the parent's invoke result. A skill can therefore carry a dozen narrow procedures while costing the roster one line. `order` in the manifest names the subskills that lead. Any subskill it omits still appears, sorted by name, after the ordered ones — so adding a subskill folder is the only step needed to publish it. ## Publishing one Make a skill and it is live: it joins the roster agents read and it opens by name from the moment it exists. Edit its description and that is the line agents see next. A skill lives anywhere under `/zero/source/`. `asset.create` hands you a valid one to fill in: ```luau asset.create("agentSkill", "waveSpawning", { folder = "skills" }) ``` A skill's scope — `builtin`, `library`, or `world` — comes from where it lives rather than from anything it declares, so a world's own skill always reads as the world's. ## Opening one ``` agent_skill {} -- the roster agent_skill { name = "scenes" } -- one skill's instructions + deps agent_skill { name = "scenes/player-setup" } -- a subskill ``` The same three through the toolbox — `skills.list`, `skills.invoke` — and from Luau via `tools.use("skills", "invoke", "scenes")`. `asset.inspect` on a skill ref returns the structured record all of them render. Skills authored in a world are the world's own, and are labelled as world content wherever they surface beside built-in ones. ## In the Inspector Skill: what the skill is for and when it applies, how many checks and verdicts it declares, its stages, a link to each asset it depends on (and a problem for each that does not resolve), its subskills, and **Open** for instructions.md.

# scene.player The scene's player setup, in one tool. ```lua scene.player() -- report: intent + prototypes (body/camera wiring) + spawns scene.player("none") -- playerless scene; removes every PlayerPrototype + PlayerSpawn scene.player("spawns") -- player per joining user; authors the default setup if missing scene.player("spawns", { body = "@builtin::avatars.minimal_player" }) -- swap the avatar scene.player("spawns", { behavior = "@builtin::controller.first_person" }) -- first-person scene.player("spawns", { camera = "my_cam" }) -- wire an existing camera entity ``` The intent is typed: `"spawns"` or `"none"`. `"spawns"` on a scene with no PlayerPrototype authors the canonical setup — humanoid body, third-person camera (OwnerOnly), a spawn at the origin — the same shape a fresh scene ships and the `player` template produces. `"none"` strips the player machinery entirely. A `roles` table (second argument) reshapes the player and implies `"spawns"`: - `body` — an entity name/id to adopt, **or** an avatar asset ref (`@builtin::avatars.minimal_player`) that repoints the body's avatar. A joining user's own avatar replaces this body socket, so reuse the canonical setup rather than hand-rolling one. - `camera` — an entity carrying a Camera component; moved under the prototype and scoped OwnerOnly so each joining user gets their own view. - `behavior` — any `cameraBehavior` component (built-in or authored) set on the player camera's `Camera.behavior`. - `prototype` — picks one when the scene has several. Setting anything saves the scene. The report names the root scene layer's `intent` and `scene` path, and lists every PlayerPrototype and PlayerSpawn the loaded layers hold. In play the scene holds each authored prototype as the captured template a joining player is cloned from; the report lists it through the spawn that names it, with `template = true`. With no scene loaded as the root layer the report carries `unloaded` in place of `intent` and `scene`, saying so and how to load one, and the setters refuse until a scene is loaded. The behavior is attached to the player camera's entity as a component of its own type, so it is tuned there: `component.get("<its type>")` on the camera entity and set its fields. A `component.add` of that type writes its values onto the attached instance instead of adding a second one.

# 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`

# inputSim Every simulated input, across every device a player might hold. Drive the world as a keyboard-and-mouse player, as a phone, or as someone with a controller — and read back what actually reached the game. ## Why all three, always A world checked only from a keyboard is a world whose keyboard experience is the only one anybody verified. That is how a game ships playable on a desktop and dead on a phone: not because the touch path was written badly, but because nothing ever drove it. So this toolbox covers the three classes symmetrically, and `device` makes a desktop session present as any of them. ## Driving the game: actions A game reads its input through the controls its input maps declare — `move`, `jump`, `sprint`. `action` drives one by name with the value the game should read, which is how a test or an agent plays a game: the same call works whatever device the player would hold, and nothing about a device — which key a map binds, where the pointer is — has to be worked out. ```lua tools.use("inputSim", "bindings") -- the live controls and their kinds tools.use("inputSim", "action", "jump") -- tap a button tools.use("inputSim", "action", "move", { 0, 1 }, 2) -- walk forward for 2 s tools.use("inputSim", "action", "sprint", true, nil, true) -- hold until `release` tools.use("inputSim", "fired") -- reported under class "driven" ``` A button takes `true`, a 1-D axis a number, a 2-D axis `{ x, y }`. The value joins the control like another device: its shaping applies, its press and release fire, and a context or a suppression that stands the control down stands it down too. `release` lets go of every driven control. The device verbs below prove the devices themselves: that a key, a pad or the touch overlay reaches the control a player would press it for. ## The loop ```lua tools.use("inputSim", "device", "touch") -- present as a phone tools.use("inputSim", "bindings") -- what controls are live? tools.use("inputSim", "layout") -- what is on screen? tools.use("inputSim", "tapButton", "Jump") -- tap it where a player would tools.use("inputSim", "fired") -- did it reach the game? ``` `bindings` first, always. Nothing is live until a component activates an input map, so an empty list means no awake component asked for input — which is a different problem from a control that fires and does nothing. `fired` is what turns a simulated input into an assertion. A capture can look identical whether or not the input landed; this cannot. Its `count` is the number of engine frames the control was active, so a tap that holds a button across two frames reads 2 for one press. To count presses, subscribe with `control:onPressed`. ## One argument shape Every verb here takes its arguments POSITIONALLY, in one order across the whole toolbox: **what** first — the button, key, stick, trigger, label or contact id — then **where** as flat `x, y` (or `dx, dy`), then **how long** in seconds or **how far spread** in frames, then **which pad slot**. Every optional argument trails. No verb wraps its arguments in an options table, so a verb's shape is known from having called its neighbour. ```lua tools.use("inputSim", "click", 0, 400, 300) -- what, where tools.use("inputSim", "mouseDown", 0) -- the same what tools.use("inputSim", "touch", 400, 300, 0.5) -- where, how long tools.use("inputSim", "padStick", "left", 0, -1) -- what, where ``` Every surface that names arguments resolves those names against the same positional list, so one call is written four ways and reaches the same slots: `click(0, 400, 300)` in Luau, `{ button = 0, x = 400, y = 300 }` through `tools.use`, `args` over `use_tool`, and `--button 0 --x 400 --y 300` in the shell. A verb whose every argument is optional takes `{}` for "no arguments at all", and a name the verb does not declare is refused with the ones it does. ## Who the input is for What this toolbox simulates is the game's input: a key, click, motion, wheel or touch reaches the scene alone. It reaches the scene whatever a person at the window has focused — a panel they clicked, a terminal they are typing a message into — and it reaches no UI widget, so a run driving the game never types into the person's panel and the person's typing never stands the run down. Every key, pointer and touch verb takes a trailing `target`: `"game"`, the default, or `"ui"`, which reaches the UI the way a person's device does. Driving a UI by its elements — clicking a button, typing into a field — is the `gui` toolbox's work. ```lua tools.use("inputSim", "key", "KeyW", 1) -- walks the player tools.use("inputSim", "key", "KeyA", nil, "ui") -- types into the focused field ``` A mouse button is written either way, index or name: `0` / `"left"`, `1` / `"right"`, `2` / `"middle"` — the same pair `Zin.state.mouseButtonDown` reads. A pad button is always a canonical position name (`"south"`, `"dpad_up"`, …), because a pad has no index a player could recognise. ## Where a point is measured from `mouseMove` and `click` take their point in physical pixels, measured in the space the input's receiver reads positions in, so `target` decides it. With `target` `"game"` (the default) the point is in the space `capture` reports: the scene's own pixels, measured from the top-left of the picture a capture returns. That is the space `Zin.state.get().mouse_position` reads a pointer back in and the space `worldToScreen` answers in, so a point read off any of the three is a point the other two take. ```lua tools.use("inputSim", "mouseMove", 300, 200) Zin.state.get().mouse_position -- 300, 200 ``` The window can hold the scene inside a rect with editor chrome above and to the left of it; the offset is those two verbs' to carry, not the caller's. With `target` `"ui"` the point is on the window: the space every UI widget is recorded in, wherever its screen draws (the window itself, or the picture a Scene panel presents). A widget's `gui.bounds` rect is in logical points, and times `ui.pixelRatio()` it is the point to aim at: ```lua local b = tools.use("gui", "bounds", "play-button") local r = ui.pixelRatio() tools.use("inputSim", "click", 0, b.centerX * r, b.centerY * r, nil, "ui") ``` `gui.clickElement` does the same by id and refuses a widget the last frame did not draw where its rect claims, so it is the call for pressing a named button. A `macro`'s ops are those same verbs, so an absolute point written for one is the point the other takes. `touch`, `touchDown` and `touchMove` place their contact on the window. `camera.viewData().viewport` states where the scene's rect sits on that window. ## Keyboard and mouse | Tool | What it does | |---|---| | `key` | Tap a key — down, optional hold, up | | `keyDown` / `keyUp` | Hold and release | | `click` | A full click, optionally at a point | | `mouseDown` / `mouseUp` | Hold and release a button | | `mouseMove` | Move the cursor to a point; the delta is what look bindings read | | `mouseMoveBy` | Move the pointer by a motion, over `steps` frames — repeat it to hold a look axis | | `scroll` | Turn the wheel | | `lockPointer` | Lock or release the pointer — mouse-look is usually gated on it | `keyDown` and `mouseDown` hold until something releases them, and a hold outlives the call that made it — including a driver task that was cancelled mid-press. `held` says what is down; `release` puts all of it down at once. ## Touch | Tool | What it does | |---|---| | `touch` | A whole tap at a point | | `touchDown` / `touchMove` / `touchUp` | One finger, by contact id — several at once | | `stick` | Push the on-screen movement stick | | `drag` | Drag across the look zone, spread over frames | | `pinch` | Pinch or spread two fingers | | `tapButton` | Tap an on-screen button **by the label a player reads** | `tapButton` aims at the button's real centre and raises when no live button carries that label, listing the ones that do. That is deliberate: a tap that silently lands on nothing looks exactly like a control that works. ## Gamepad | Tool | What it does | |---|---| | `pad` / `padOff` | Connect or disconnect a simulated pad | | `padPress` | Press a button — down, optional hold, up | | `padDown` / `padUp` | Hold and release | | `padStick` | Push a thumbstick | | `padTrigger` | Squeeze a trigger to an analog travel | Buttons are named by POSITION, never by the letter printed on them: `south`, `east`, `west`, `north`, `left_shoulder`, `right_shoulder`, `left_trigger`, `right_trigger`, `left_stick`, `right_stick`, `select`, `start`, `guide`, `dpad_up`, `dpad_down`, `dpad_left`, `dpad_right`. The same call works whatever pad is connected. The name passed to `pad` decides which legends a prompt draws, so connecting `"Sony DualSense Wireless Controller"` is how a scheme's PlayStation prompts get checked. ## Reading back | Tool | What it answers | |---|---| | `device` | Which class the session presents as — and forces one | | `bindings` | Which controls are live, and which map contributed each | | `layout` | What the on-screen controls are actually drawing | | `fired` | Which controls were active since last asked, for how many engine frames, and from which device | | `held` | What the session is holding down right now, across every device | `layout` is how a phantom control is caught: a label on screen that no entry in `bindings` explains is a button nothing consumes. `held` is how a leftover hold is caught: keys, mouse buttons, fingers, the stick and connected pads in one answer, with `atRest` true when the session holds nothing. A control that keeps firing after a take ended reads here as the key that is still down, so recovering from it is a `release` rather than a guess at what was pressed. ## Replaying a sequence A run of events worth repeating becomes a `.inputMacro` asset, so the same sequence drives the same world on every later check. | Tool | What it does | |---|---| | `recordMacro` | Capture real input for `duration` seconds | | `saveMacro` | Write an event list to a `.inputMacro` asset | | `macro` | Play an event list back through the input surface | | `listMacros` | Every `.inputMacro` asset in the project | An event is `{ t, op, arg }`. `op` names one of the verbs above that drives input and `arg` carries that verb's arguments — the value itself when the verb takes one, a table keyed by the verb's parameter names when it takes several — so every device class a verb reaches is a device class a macro reaches. A `touchDown` event's `arg.id` is the name its contact goes under, and the `touchMove` and `touchUp` events carrying that `id` drive it. ```lua tools.use("inputSim", "macro", { { t = 0.0, op = "pad" }, { t = 0.4, op = "padDown", arg = "south" }, { t = 0.6, op = "padUp", arg = "south" }, { t = 0.8, op = "touchDown", arg = { x = 714, y = 984, id = 1 } }, { t = 1.2, op = "touchUp", arg = 1 }, }) ``` `recordMacro` reads what every device is HOLDING each frame — keys, mouse buttons, pad buttons, the pad itself, fingers on the glass — and what each one is MEASURING every `sampleInterval` seconds — mouse motion, the wheel, thumbsticks, triggers, a sliding contact. `release` is the reset between takes — it lifts every contact, drops the on-screen stick, releases every held key and mouse button, and disconnects every connected pad, so one take never leaks a held control into the next. It reports how many of each it let go, and `held` reads the same set before or after.

# camera.bindToPlayer Make the joining player use this camera in play by binding it to the active scene's PlayerPrototype: it re-parents the camera into the prototype subtree (so it is cloned with the player on spawn), sets the prototype's `camera` role to it, and detaches + disables any displaced camera (parked idle, not deleted). Pair it with a static behavior (`@builtin::controller.menu`) for a fixed PS1-style camera the player moves within, or a follow behavior for a tracking one. A player-less scene has no prototype — there the rendering camera is simply the highest-priority one.

# capture Render a screenshot of the scene from anywhere — an explicit position, a named camera station around an entity, a specific camera, or a mirror of the active viewport — and a grid of them in one image; or read back the frame the engine actually presented, exactly as the player saw it. The picture of the machine's own camera (a webcam, a phone camera) is `cameraInput`, covered by the `topics/camera-input` guide; this toolbox photographs the scene. Every source but `presented` and `window` (see [The presented frame](#the-presented-frame)) spawns an ephemeral offscreen camera, triggers one render, and returns the picture as an image value — returned from a tool, a probe or `execute`, it reaches you as a picture, with what the capture has to say about it (its size, the frame it is of, the cells of a grid) beside it. A PNG is written as well when you pass a `path`. The convenience wrappers render from a world-space position (with optional look-at), frame a target entity, mirror the active viewport camera including its layers, or render a specific camera by name from its exact authored pose and settings (`fromCamera`, with optional per-capture overrides that never touch the camera); they run as a background task that waits for the GPU readback and cleans up — which only works in play mode, since the task scheduler ticks there. The low-level oneshot spawns the camera, triggers one render, and returns the handles synchronously — the only path that works from edit mode, and the one the `capture` MCP tool is built on. A cleanup tool clears the capture output directory. ## Where a shot lands A capture given a `path` writes the PNG there and says where those bytes stand, beside the path it wrote: `durable` is true when the PNG is on the path it names and the next session reads it back, and false when the play shadow took it (live in this session and discarded on a guarded play-exit unless kept, with `survivesRestart` saying whether a restart keeps it and `playShadow` naming the routes that keep it) or when the path is in `/source/tmp/` scratch (gone when the engine stops or restarts, with a `warning` opening `SCRATCH`). Every capture carries it on the picture it returns as `durability`. Shooting gameplay means shooting in play, which is exactly when a `/source` destination is the shadow's, so this is the field that says whether a shot survives leaving play — `vfs.promotePlayShadow(path)` keeps it with play still running. ## The presented frame `source = "presented"` reads back the frame the engine presented to the player instead of rendering one. Nothing is drawn for it: it is the frame the viewport drew, copied out as it was finished, after the post chain, the upscale, the display transform, the content UI and every post-scene pass, at the size it was composed at. So it is the one capture whose temporal antialiasing, motion blur, camera motion, auto exposure and render scale are the real ones, and the source to verify any of them with. `source = "window"` is the whole presented surface (the window, the browser canvas or the headless framebuffer) with everything painted on the window: the editor's chrome and every window-level UI pass. Where no Scene panel shows the viewport, the two are the same image: the scene fills the window and the editor's chrome is painted over it, and each such frame says so in its `note` (a cell of a `presented` capture, a frame `renderer.presentedFrames` returns) for as long as that chrome is on the window. Show the Scene panel (the editor toolbox's `layout` tool) to read the game view alone. `frames = n` (1-32) takes n presented frames, one every `interval` frames (default 1: consecutive), at real gameplay speed, and lays them out as a grid in the order they were presented. Each cell states what its frame was drawn with: `frame` (the renderer's count, so consecutive frames carry consecutive numbers), `rasterWidth` / `rasterHeight` (the size the scene rasterized at), `compositeWidth` / `compositeHeight`, `renderScale` and `scaleMode`, `jitter` (the projection's sub-pixel offset in raster pixels), `exposure` (what the display transform multiplied the scene by) and `historyReset` (the viewport camera's first frame, a cut, or a hand-over to another camera). A single frame comes back at the size it was presented at; `width` / `height` resample it, or size the sheet a grid is laid into. A presented frame is drawn through the camera, layers, lens and chain that drew the screen, so the options that aim, light or shape a rendered frame (`camera`, `entity`, `position`, `viewpoint`, `fov`, `pass`, `renderLayers`, `postProcessing`, `environment`, `mode = "collage"` and the collage axes) are refused for it; aim the camera holding the viewport instead, or take an offscreen capture. A source that renders refuses `frames` and `interval` the same way, since they count presented frames. `deterministic` and `clearAir` hold the clock and the air for the whole engine, so they reach the presented frame too. A diagnostic buffer of the viewport itself, under its real camera motion and history, is read the same way: put the camera holding the viewport on that pass (the camera toolbox's `set` with `debugPass = "motion_vectors"`) and take `source = "presented"`. The viewport's velocity buffer is the frame's travel against the frame before it, so a turning camera reads in it, where an offscreen `pass = "motion_vectors"` holds the objects' own motion alone. The `motion_vectors` picture is a visualization: each channel is the square root of 100 times the stored displacement's size on its axis (R horizontal, G vertical, B the length), the sign is dropped, and R reads 255 from `width / 200` px of horizontal travel, G from `height / 200` px of vertical travel (raster pixels). The rendering guide's "What the `motion_vectors` picture encodes" states it in full. The same frames, with their pixels as `TextureCpuHandle`s, come from `renderer.presentedFrames({ scope = "viewport" | "window", frames, interval })`, and `velocity = true` adds the viewport's velocity buffer from each frame as numbers (`frame.velocity`, a `VelocityField`): each pixel's travel in pixels per frame, x right and y down, signed and unclamped, read with `field:at(x, y)`, `field:row(y)` and `field:region(rect)`. ## The feed Every capture publishes the frame it made to `@builtin::tools.capture.feed`, so a viewer can show the newest one without taking a shot of its own. `Feed.latest()` is that frame: a `src` an image widget draws (the render target's guid while the engine still holds it, the written path when the capture produced a file), with the size, the source and the pass beside it. `Feed.watch(fn)` calls `fn` as each new frame arrives. The editor's Capture panel is built on it, and the `watch` layout puts that panel under the scene. A viewer calls `Feed.shown()` to say it is drawing frames from the feed. From then on the feed holds the newest capture's render target so the picture stays up after the capture that made it is finished with it, and lets it go when a later frame replaces it; `Feed.shown(false)` stops. With nobody drawing, a capture's target is released the moment its capture is done, and nothing is held. ## Viewpoints A **viewpoint** names where the camera stands and which way is up in the image: `front`, `back`, `left`, `right`, `top`, `bottom`, `iso`. It replaces reverse-engineering a `{yaw, pitch}` pair, and names the seven stations that answer "is this built correctly". `basis` says which axes the name is measured against, and it is **required** alongside a viewpoint when a subject is being framed: - **`"local"`** — the subject's own axes. `front` is the side it faces however it is turned, and the frame is fitted to its own extents rather than to the world-axis box around them. - **`"world"`** — the world axes. `front` is whatever faces world +Z. For a crate turned 40 degrees those are different pictures, so the call says which one it wants. Framing a world position rather than an entity, both values mean the world axes and either is accepted. ## Passes `pass` names **what** the frame holds, by name. `"final"` is the lit picture; a built-in diagnostic pass (`albedo`, `normal`, `depth`, `motion_vectors`, …) replaces the shaded image with the buffer behind it; and any capture **view** a render feature has published (`renderer.captureView.list()` enumerates them) is selectable by its own name. A name that is neither is refused against the list of what is registered. The name carries two decisions: which channel the camera renders, and which render layers that channel is legible under. A diagnostic buffer is read flat, so the sky behind it and the post-process chain over it come off; a view that draws its own geometry declares the layers it needs and is rendered under them. Stating `renderLayers` answers that second question yourself, for the one call. `passes` spreads the same vocabulary across the cells of a collage, and each cell reports the layers it was read under. ## Projection `projection = "orthographic"` covers a fixed world-space height at every distance. Parallel edges stay parallel and two equal-size objects at different depths cover equal pixels, which is what makes it the projection to check a shape or compare sizes under — a perspective frame's convergence hides both. `orthoHeight` sets that height; omit it and it is fitted to what is being framed. Any camera can be orthographic, not only a capture: it is a field on the `Camera` component. A capture through a camera the scene holds (`fromCamera`, `viewport`, and `source = "camera"`, `"screen"` or the scene camera) takes that camera's `projection` and `orthoHeight` with the rest of its lens, and its lighting `environment` with an `isolated` camera's background and ambient; a `projection` or `orthoHeight` stated for the call stands over the camera's. ## Which entity a capture framed A capture names an entity, or a camera to render from, by its id or by its name. A name exactly one entity carries means that entity. A name several entities carry means none of them on its own, so the capture is refused, and the refusal says how many entities carry it and names each by name and id; pass the id of the one you mean. A player is drawn as its body, so a capture of a player identity (an entity carrying `UserIdentity`, named `user`) frames the body its `avatar` names. A player prototype is absent from the live scene while it plays, so its name reaches no entity there: a capture of that name is refused, and the refusal names each player spawned from the prototype, by its identity and its body, with their ids. Every cell of a capture aimed at an entity reports what it framed under `subjects`: one `{ id, name }` per subject, carrying `playerIdentity = { id, name }` when the picture is the body of the identity the capture named. ## Isolate Framing a subject fits the **frame** to it; it does not decide what is drawn in that frame. Everything else standing there is still rendered — and a scene's default player spawn sits at the world **origin**, which is exactly where a prop usually gets built, so a tight shot of a new prop at the origin can come back showing the player instead. `isolate = true` draws the subject without the other geometry. It changes exactly that — lighting, sky and post-process are untouched, so a final frame stays a final frame, and a flat read of the geometry is a `pass`. Excluded geometry still casts shadows onto the subject and still bounces light into it, the same as any render-layer exclusion. In a `setups` collage each cell shows the entity its own set-up frames alone, and a set-up can state `isolate` for its own cell; a set-up aimed from a position or a camera frames no entity and is refused when it is to be isolated. ## The tone the frame carries Every capture reports `luma` on its cell record: the tone of the pixels it just encoded, in code values on the 0-255 scale the image is delivered at. ``` luma = { pixels, min, max, mean, p1, p5, p50, p95, p99, span, spread, crushed, clipped } ``` `span` is `max - min` — the whole range, a single stray pixel included. `spread` is `p95 - p5` — the range the body of the picture occupies, which is the reading that answers whether a shot is legible. `crushed` and `clipped` are the shares of the picture sitting at code 0 and at code 255. A frame's mean cannot answer that question: half a picture at 40 and half at 200 carries the same mean as a flat field at 120, and only one of them has anything in it. A subject can be modelled, shaded, lit and drawn and still arrive inside a handful of code values, in which case the image looks like a wash and every other field here reports success. `spread` is the number that says so. The same reading is available anywhere pixels are: `cpu:tone()` on a `TextureCpuHandle` — from `renderer.texture.readback(rtHandle)` on a capture handle, from a render-to-texture camera's target, from a CPU canvas — and `cpu:histogram()` under it, the per-code-value counts for luminance and for each of the three channels, which every reading above is a sum over. `cpu:histogram(x, y, w, h)` counts one rectangle of the picture alone, so a reading per region of the frame — how much of each cell is dark, lit or a given colour — is one call per region. `cpu:flip(reference)` holds the picture against another of its size by NVIDIA FLIP, the comparison a golden is judged by; `@builtin::modules.postMetrics` carries it with the golden runner, the pixel metrics (banding, temporal variance, halo overshoot, ring energy, hue drift) and the scopes, each reading a presented frame or a capture result as it comes. ## The grade an offscreen frame carries A capture spawns its own camera and renders offscreen, and the post-process chain runs over that frame the way it runs over the viewport's — for a station, for an orbit around an entity, for a named camera, for a render-to-texture camera alike. Inside an effect, `engine.view_proj` / `engine.prev_view_proj` / `engine.inv_view_proj` are the camera THAT render was drawn from and `engine.resolution` is that target's own size, so a grade that reconstructs world space from scene depth reconstructs against the station and lens the capture asked for. That makes an offscreen capture an oracle for an authored look: it photographs a chosen station and lens without taking the on-screen camera away from whoever else is driving the scene, and moving that camera somewhere else does not move the frame this one returns. An offscreen render keeps no view history of its own, so `engine.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 a camera's motion (and the blur and temporal reconstruction built on it) is read from the presented frame: `source = "presented"`. 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. `render(ctx)` itself runs once a frame for the main viewport, so `ctx.viewport` is the window's size rather than the captured target's; a feature sizes its scratch with `screen = true` and lets the engine track the drawn target. `postProcessing = false` takes the chain off the frame a capture returns, and with it the film a render feature draws — the flare and grain, a short focus's bokeh, the shutter's smear. It is a `Camera` property rather than a render layer, so `renderLayers` does not reach it. A capture naming a diagnostic `pass` drops the chain by default, so the buffer comes back as the value it encodes. ## Collage `collage` renders a grid of frames into one image. What varies across its cells is what you pass — `setups` for a cell per whole camera set-up, `viewpoints` for a cell per view of one subject (or `"sides"` for all six), `passes` for a cell per render pass, `duration` for a cell per sample over time. A collage with no axis is refused rather than quietly sampling time. `setups` is the shot list: each entry is written the way a single capture is aimed (`source`, `camera`, `entity`/`entities`, `position`, `lookAt`, `rotation`, `distance`, `angle`, `margin`, `viewpoint`, `basis`, `projection`, `orthoHeight`, `fov`, `near`, `far`), plus `label` to name the shot on the result, and a field an entry leaves out comes from the collage's own options. `setups` and `viewpoints` are one axis — both say where the camera stands — so a request names one of them; either crosses with `passes` into a 2D grid, a row per station and a column per pass. The shot most of this exists for is one call: six sides of one object, each an orthographic elevation, nothing else in frame. Authoring the cameras themselves — creating, listing, aiming, framing, and setting their parameters — is the `camera` toolbox.

# controller.side_scroller Side-scroller camera controller. Fixed-offset camera that follows a target from the side (classic 2.5D perspective). Maintains a constant position relative to the target and looks slightly ahead. Public fields: `target`, `centerX`, `centerY`, `centerZ` (fallback), `offsetX`, `offsetY`, `offsetZ` (camera offset from target), `lookAheadX`, `lookAheadY`, `lookAheadZ` (look-at offset from target). Controls: none — follows the target automatically. ```luau entity(id).component.add("controller.side_scroller", { target = "player", offsetX = -8, offsetY = 3 }) ```

# controller.first_person First-person camera controller. Positions the camera at an entity's location (+ eye-height offset) and rotates with mouse look. It resolves its follow target in this order: its own `target` field, then `Camera.follow` when `target` is empty, then the static `posX/posY/posZ` when neither is set. `Camera.follow` is the standard slot every behavior reads, so a first-person camera selected through `Camera.behavior` tracks whatever `Camera.follow` points at — the correct source for a player camera, whose body id is minted fresh per spawned clone and cannot be authored onto the behavior ahead of time. Public fields: `target` (entity id to follow, empty = use `Camera.follow`, then `posX/Y/Z`), `posX`, `posY`, `posZ`, `offsetX`, `offsetY` (eye height above target origin), `offsetZ`, `sensitivity`, `minPitch`, `maxPitch`. Controls: pointer lock to rotate (yaw + pitch), which the standard scheme takes while the world runs. With the cursor handed back, an RMB drag rotates too — for as long as nothing else in the world binds the right button. ```luau -- Track the Camera's follow target (set Camera.follow, no id authored here): entity(id).component.add("controller.first_person", { offsetY = 1.6 }) -- Or a static camera at a fixed position: entity(id).component.add("controller.first_person", { posX = 0, posY = 1.6, posZ = 3 }) ```

# camera.set Change one or more camera parameters in a single call: `fov`, `near`, `far`, `priority`, `debugChannel`, `renderLayers` (a layer-name spec like `"all !ui"`), `behavior` (a `cameraBehavior`-tagged component), and `follow` (the entity the behavior tracks). Pass no camera to change the active one.

# 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.

# animation toolbox Test-play animation clips on skinned bodies. `animation.play` is the quick way to see a `.animation` clip on a character without authoring a component: it resolves the clip and the entity, finds the SkinnedModel to drive, plays the clip through the canonical AnimGraph, and tears everything down when the window ends. ```lua animation.play("A_Walk_F_Masc", "hero") -- play once, auto-cleanup animation.play("Idle", "hero", { offset = "5s" }) -- wait 5s, then play animation.play("Wave", "hero", { loop = true, duration = 8 }) -- loop for 8s ``` It plays while the world runs (play mode), the same as any component-driven animation. Durable playback belongs on a `ClipPlayer` component; this tool is for throwaway testing and leaves no component behind. On a body that already animates (a ClipPlayer, the character controller) the clip overrides for the window and the body's own animation resumes untouched afterwards — the tool reuses the running graph and restores its output on cleanup.

# inputAuthor Write control schemes whose every control already carries keyboard, gamepad and touch. ## Why a toolbox for this A binding must cover all three devices — that rule is what stops a world shipping playable on a desktop and dead on a phone. A rule like that only survives if obeying it is cheap; otherwise authors reach for a canonical name that half-fits, or bind the keyboard and mean to come back. So the catalog fills in the common controls. Ask for `jump` and get Space, the south face button, and a lower-right touch button, already complete. Ask for six controls and get a whole scheme in one call. ```lua tools.use("inputAuthor", "map", "shooter", { "move", "look", "attack", "aim", "reload", "jump" }) ``` Where each control lands is a judgement about the device rather than a translation of the keyboard binding: `sprint` is a stick click on a pad because that is where a thumb already rests, and a lower-right button on touch for the same reason. `aim` is a trigger, not a face button. ## Overriding A control is either a catalog name or a table naming it with overrides: ```lua tools.use("inputAuthor", "map", "racer", { "move", "brake", -- a sprint, under the name this game calls it { name = "boost", from = "sprint", label = "Nitro" }, { name = "horn", -- unknown: supply all three kind = "button", kbm = '{ B.key("KeyH") }', gamepad = '{ B.padButton("north") }', touch = '{ B.touchButton({ zone = "right-lower" }) }' }, }) ``` `from` is for the common case where a game's control IS a canonical one under its own name — a racer's boost is a sprint, a shooter's fire is an attack. It keeps the control named what the game calls it without retyping three bindings. A name the catalog does not know, and that does not say what it is based on, needs all three classes from you. It cannot guess where a control nobody has named before belongs on a pad or a phone screen, and guessing would produce exactly the half-right layout this whole model exists to prevent. Nothing is written until every control resolves, so a scheme with one unknown control names it rather than leaving half a map on disk. ## Tools | Tool | What it does | |---|---| | `map` | Write a scheme: the map folder and every control it declares | | `control` | Add one control to a scheme that already exists | | `catalog` | List the controls that come pre-filled, and what each is bound to | | `check` | Report every problem across a scheme's controls, without activating it | `check` reports all faults at once and activates nothing, so a scheme can be inspected before a world runs it. To drive the result on each device, use the `inputSim` toolbox.

# appearance Change how entities look across a whole hierarchy in one call — hand it a material, a mesh, or a shader and a scope, and it finds the right entities and fields and applies it. The common situation is that you have a material and a rough idea of where the thing lives, and want the assignment done across the whole subtree at once. Batch-assign a material (by name or ref) to every material field under a scope, swap the mesh on the Model / SkinnedModel entities under a root, or swap a material's shader while preserving its texture links and property values by canonical role. Targeting is schema-driven: for the material and shader operations, each targeted entity's components are enumerated and every component — builtin or user-authored — that declares a public material asset-ref field participates. The scope is one entity, a list, or the whole scene by default; options can stop at the first match or filter candidates by name pattern. This is the hierarchy-aware assignment layer — creating materials and editing registry-wide properties live on the lower-level `Material` surface, and per-ref edits live on the resolved asset refs themselves.

# wld.play Enter play mode. First auto-snapshots the current edit state so returning to edit can restore it, then starts scripts and physics so the scene runs live. This is the counterpart to returning to edit — play the scene to see it in motion, then rewind to the pre-play snapshot.

# multiplayer guide Reference docs surfaced through `guides { path: "..." }` and `man multiplayer`. The content lives in `guide.md`; this README is just the asset summary.

# third_person_follow Over-the-shoulder camera. The rig hangs off a shoulder pivot beside the follow target and sits back along the heading it looks down, so framing is authored as offsets in space. The offset is measured in the camera's own frame, so the subject holds the same spot on screen whichever way it turns. The rig owns that heading. Binding a follow target aims it along the subject's facing, 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 facing a bind reads comes from the body's own declaration of which way it faces (`Locomotion.forward`), so the rig lands behind a `+Z`-forward Synty body and a `-Z`-forward glTF body alike. A target that declares nothing falls back to its transform forward. Public fields: `follow`, `shoulderOffset` (x across the frame, y up it, z along the heading), `verticalArmLength`, `cameraSide` (0 = left shoulder, 1 = right), `cameraDistance`, `minDistance`, `maxDistance`, `aimYaw` (the world heading the camera looks along, in radians about +Y from world +Z) and `aimPitch` (how far it tilts off level), `damping` (per-axis time constants in seconds), `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; on pitches the other way), `zoomSpeed`, `minPitch`, `maxPitch`. Writing any framing field re-poses the rig immediately with damping bypassed, so edit mode shows the framing play mode will produce. `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 look (play mode requests pointer lock, so no button is held), scroll to adjust `cameraDistance`. ## The control scheme `inputMap` is the scheme the rig answers to. It reads two controls from it, each on its own: - `look` — turns `aimYaw` and `aimPitch`. - `zoom` — steps `cameraDistance`. A map declaring only one of the two gives the rig that one and keeps the other where it is; the rig logs which control the map left out, and names the map. A field naming no map leaves the rig answering nothing, and logs that too. Point `inputMap` at a world's own map to rebind looking and zooming without touching this component. **Repointing it on a live rig takes on the next frame.** The map the rig was holding is stood down — its controls leave the live set and its touch buttons leave the screen — and the map named here is taken up in its place: ```luau entity(rigId).component.get("third_person_follow").inputMap = asset.resolve("mygame.input.cinema", "inputMap") ``` Assigning the map the rig already holds changes nothing, so re-stating the scheme leaves the live set where it is. While gameplay is paused the rig holds no map at all — the editor's camera has the controls — and it takes whatever `inputMap` names when play resumes. ```luau local cam = entity(id).component.get("Camera") cam.behavior = asset.ref("@builtin::controller.third_person_follow", "component") entity(id).component.get("controller.third_person_follow").cameraDistance = 3 ``` ## 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.third_person_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. Assigning the slot attaches the rig to the camera's entity, so it is tuned on that instance: `component.get("third_person_follow")` and set its fields. A `component.add("third_person_follow", { ... })` after the assignment, in a scene `build.luau` or anywhere else, writes its values onto the instance the slot attached and returns it, so the Camera keeps the one rig it knows about. In a `build.luau` that add is the way to tune the rig: its values are saved with the scene and the loaded rig holds them. ## 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.

# 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`

# camera.lookAt Aim a camera at a world position or at another entity. Pass a single target to aim the active camera, or a camera plus a target to aim a specific one.

# topics/input guide Reference docs surfaced through `guides { path: "..." }` and `man topics/input`. The content lives in `guide.md`; this README is just the asset summary.

# camera.frame Position and aim a camera to frame a target entity in one step: place the camera at a distance from the target on a yaw/pitch angle, then look at it.

# zinput zinput is the engine's Luau-side input framework and the **only** public Luau input API. It sits on top of the internal `__zero_input.*` FFI (snapshot, events, frameId, pointer-lock, simulate\*) and builds the control model above it in pure Luau: control assets grouped into maps, the live set of activated maps and its per-frame dispatch, per-device-class binding evaluation, the on-screen touch overlay, gesture recognisers, and the tick coordinator that advances all of it. The top-level module re-exports every submodule under a single table and owns the tick coordinator (`Zin.tick`) plus the unified handle namespace for `Zin.disconnect`. ## The control model A control is a `<name>.inputBinding/` folder inside a `<name>.inputMap/` folder. It carries a `label`, a `kind` (`button` / `axis1` / `axis2`), and a binding for each of the three device classes — `kbm`, `gamepad`, `touch`. A class set to `false` is a deliberate refusal. Nothing is live until something activates a map. A world activates the map it needs and subscribes to the controls it uses: ```luau public = { inputMap = Field.assetRef("inputMap", "@builtin::inputMaps.default", NoSync, "The controls this component reads."), speed = Field.number(5, NoSync, "World units per second at full deflection."), jumps = Field.number(0, NoSync, "How many times the jump control was pressed."), } local controls function awake() controls = public.inputMap:activate() controls.move:onInput(function(v, dt, active) if not active then return end local p = public.entity.position local step = public.speed * dt public.entity.position = { p.x + v.x * step, p.y, p.z - v.y * step } end) controls.jump:onPressed(function() public.jumps += 1 end) end function onDestroy() public.inputMap:deactivate(controls) end ``` What `activate` returned is this holder's claim on the map, and handing it back to `deactivate` disconnects the subscriptions made through it while the map stays live for whoever else holds it. Every activation reads the map's binding children again, so a control edited since it came up reaches the running set and the holders already standing on it. Several maps are live at once and compose: the player's surface is the union of them, so picking up a gun adds Fire beside the movement stick and dropping it takes away Fire alone. A map declares the `group` its controls belong to and the groups it `suppresses` while live, which is how a car stands walking down without owning the walk map. `Zin.scheme` holds the live set and runs the dispatch. `man topics/input` is the full walkthrough. ## Exports Subsystems (re-exported from the corresponding submodule): - `M.scheme` — the live set of activated maps and their per-frame dispatch. - `M.bindings` — binding constructors, evaluators, and the legend formatter. - `M.state` — per-frame snapshot reads plus held-time and synthetic repeat. - `M.events` — frame-event readers and `on`/`off` edge subscriptions. - `M.input` — `onBegan` / `onEnded` / `onChanged` / `onTextInput` listeners and `lastInputType()`. - `M.context` — the input context stack. - `M.chords` — sequence and simultaneous recognisers. - `M.gestures` — tap / double-tap / long-press / swipe / pinch / pan recognisers. - `M.touch` — per-finger contacts and `touch.*` listeners. - `M.surface` — touch-vs-kbm device-class classification. - `M.gamepad` — connected pads, family, and button legends. - `M.virtual` — the per-tick virtual-control store on-screen controls write. - `M.touchControls` — the on-screen touch overlay (stick, drag-look zone, buttons) and the API that adds a control while a world runs. - `M.emulation` — the map-driven key-emulation floor. - `M.pointer` — pointer-lock mutations. - `M.back`: Back, claims a world holds while Back means something to it, and `press` to issue one. The engine routes each press to the topmost dismissible overlay, else the most recent claim, else the platform's own default (an Android app moves to the background). - `M.test` — synchronous test-input helpers. - `M.controllers` — `free` / `orbit` / `fps` controller factories. - `M.profile` — binding profiles, saved as JSON and loadable from `@builtin::profiles.<name>`. - `M.rebind` — interactive binding-capture sessions. - `M.conflicts` — the binding-collision reverse index. - `M.autoTick` — the opt-in per-frame tick worker. - `M.utils` — pure helpers. - `M.actions` / `M.axes` / `M.map` — the name-registered action, axis and map registry. A control asset replaced them; each carries its own `--!deprecated` message naming the replacement. Tick coordinator: - `M.tick(dt: number?)` — advance every stateful subsystem. - `M.lastTickAt() -> number` / `M.lastTickFrameId() -> number` — test/debug accessors. - `M._resetTickGuard()` / `M._resetLiveArmed()` — test-only. Handles: - `M.disconnect(handle: number) -> boolean` — accepts any Zin handle (input listener OR action handler) and routes to the right `disconnect`. Constants: - `M.VERSION` — the library version string. - `M.SPEC` — the VFS path of the architecture spec. - `M.ROADMAP` — the VFS path of the roadmap document. ## Raw state `Zin.state` and `Zin.events` are the snapshot and event layer that bindings evaluate against — the layer the emulation floor, rebind capture, and dev tooling operate on directly. Gameplay code subscribes to controls; polling `Zin.state.keyDown` from a world makes that world keyboard-only and draws the `engine-raw-key-poll` diagnostic at author time. ## Notes - `Zin.tick(dt)` is the single hub that advances every stateful subsystem each frame. Stateless features (state / events / utils / bindings / context) work without ticking; the live binding dispatch, axes, chords, subscriptions, held-time, repeat, and rebind are dormant until something ticks. - The first real input read arms the input system on its own: it starts `Zin.autoTick`, so dispatch, the touch overlay and the emulation floor advance with no host component and no explicit tick call. A host script component can still call `Zin.tick(dt)` from its `update(dt)` hook, and `Zin.autoTick.start()` remains available directly; both stay idempotent alongside the self-arming path. `Zin.test` engagement suppresses the arming so suites keep explicit control over tick cadence. - Same-frame re-tick dedup: `__zero_input.events()` is non-draining, so running the observe-and-dispatch pipeline twice in one engine frame would double-fire subscribers. The frame-id guard in `M.tick` and the autoTick worker keeps this from happening. - The handle namespace is unified: every handle from `Zin.input.on*` and `Zin.actions.bind` is drawn from a single counter, and `Zin.disconnect` routes any of them to the correct `disconnect`. `Zin.input.disconnect` and `Zin.actions.disconnect` are also wired to the unified dispatcher. - Library users can either grab the whole table (`require("@builtin::modules.zinput")`) or pull one submodule (`require("@builtin::modules.zinput.state")`). - `__zero_input` is an internal namespace — reachable from Luau but hidden from `/docs/api/`; production game code goes through Zin.

# 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), `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.

# 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`, `destroy`). - You want to expose configurable fields to designers / `entity.spawn` callers / scenes without code changes. If you need a global service that outlives any entity, use a `.service` instead. If you need a stateless pure-function library, use a `.module`. ## 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 beside it on the fixed timestep; `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`, `destroy`) 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 service, 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. Update every `component.add(<oldname>, ...)` call and the scene JSON entries that reference it. ## Related types - `.module` — pure-function library, no per-entity state. - `.service` — globally-scoped behavior, 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`.

# assetType (asset type) The `assetType` is the **type of types** — the meta-type that every `<typename>.assetType/` folder is an instance of, including itself. A `<typename>.assetType/` folder declares a new asset type called `<typename>`: its on-disk shape, its identity rule, and the structural contract instances of the type must satisfy. This type is self-hosting. `assetType.assetType/` conforms to its own schema — a `type.yaml` plus a `README.md` — so the type system is defined in terms of itself rather than a hidden engine special-case (the same way a C compiler is written in C). Every asset in the engine, including each `.assetType` definition, now resolves to a registered assetType: a `.material` resolves to `material.assetType`, and `material.assetType` resolves to `assetType.assetType`, which resolves to itself. ## Why it exists Assets used to be mapped to their type *implicitly*, by matching the folder suffix against a name-keyed registry. That match carried no stored link: nothing on a `Foo.material` recorded *which* `material.assetType` it was written against, so two worlds shipping a same-named type could silently disagree. Making `assetType` a real, resolvable asset closes that gap — the link from an asset to its type is now an explicit reference pinned in the asset's `.refs` sidecar (`via: "asset_type"`), exactly like every other dependency. ## Where it lives - Source: `/zero/source/.../<typename>.assetType/` - Identity: `<typename>` (the `.assetType` suffix strips from the identity; the folder retains the suffix on disk). - Folder shape: - `type.yaml` — the structural spec the validator reads. **Required.** - `README.md` — type-level documentation. **Required.** - `about.yaml` — three short plain-language texts saying what the type is for (see "Say what your type is for" below). Every built-in type ships one. - `behavior.luau` — optional behavior/scaffolding module. - `template/` — optional canonical placeholder body the templater copies into a new asset of the type. - `<name>.module/` — optional shared code the type ships to **every instance** of it, conventionally `shared.module/` (see below). ## Shared code its instances reach (`asset.containing` + `.modules`) A type ships code every instance of it uses by declaring the modules in its own `behavior.luau`, as a map of tracked `require`s: ```luau -- inside <typename>.assetType/behavior.luau M.modules = { shared = require(".shared") } -- the sibling shared.module/ ``` An instance reaches it with one location-independent line: ```luau -- inside any foo.<typename>/init.luau local api = asset.containing(__FILE__).modules.shared ``` `__FILE__` is the chunk's own VFS path; `asset.containing` resolves the calling asset, and `.modules.<name>` follows the instance's pinned `typeRef` (the type's **guid**, recorded in the instance's `.refs` as `via = "asset_type"`) to the module its type declares. Because resolution follows that link rather than a category name, the same line in an instance of a different type resolves to that type's module, and identically-named modules in two types never collide. `asset.typeRef(target)` reads that guid for any target; `asset.resolve(asset.typeRef(target))` gives the full `AssetRef<assetType>`. See `assetTypes/README.md` § "Shared code: type-owned modules" for the full mechanism. ## How to create one Authoring a new asset type is just creating a `<typename>.assetType/` folder under `/zero/source/`. The engine registers `<typename>` as a category **the moment the folder is written** — no engine restart, no Rust change, and it works for built-in (`@builtin/assetTypes/`) and user-defined types identically. Once `<typename>` is registered, `<name>.<typename>/` folders are recognised as assets of that type and resolve their `asset_type` link back to this definition. Creating a `<name>.<typename>/` folder *before* its `<typename>.assetType/` exists fails fast: the suffix has no registered type, so the folder can't be a valid asset. Author the type first. ## Say what your type is for (`about.yaml`) Every type carries a small `about.yaml` beside its `type.yaml`, and the README you write for the type documents how to author an asset of it. Both are required, and both are meant to be written as you make the type, while you know what it is for. `about.yaml` answers three questions in plain words, one key each: ```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. Name the whole range of things it covers, in the words a game maker would use for them. - **`function`** is how it works today: what an asset of the type is made of and what the engine does with it. - **`usage`** is what a person or an agent does with one: how they pick it, place it, call it, or hand it to something else. A few habits keep it useful. Write about the type as a whole, so the text holds for every asset of it, and leave out any one asset's name. Say what the type does right now. Stay with words a game maker would say, and mention a file name or a call only where a game maker would know it. The limits are in bytes: `intent` 800, `function` 1500, `usage` 1200. Each is generous, and a few sentences usually say it. Why bother: agents working in a world read these three texts to learn what your type is for, so they reach for it when it fits and use it the way you meant. People searching find the type by what they want to make, and the three texts let them tell at a glance whether it is the one they need. `asset.create("assetType", name)` scaffolds the file with all three keys, and a README whose first paragraph is a `[[describe]]` note. Each value starts with `[[describe]]` followed by a note on what to write; replace the whole value with your own words. A value or a README line that still starts with `[[describe]]` counts as not written yet, which the push's check of each asset it carries against its type reports, naming the file. A type you leave as it was is not asked, so older types stand. Many types at once, or one you would rather not write by hand: `services describe` lists the types that lack these texts and what filling them costs, and `{ mode = "all" }` writes them into the same files, where you edit them like anything else you wrote. Where the texts show up: `asset.inspect("<typename>", "assetType")` returns them as `detail.about`, `assetTypeRef:getAbout()` returns them as a table with the keys `intent`, `function` and `usage`, the type's `types/<typename>` guide page ends with an "About this type" section holding them, and the Inspector's Type section lists them at the top. The guide page is this README plus that section, so a README that explains the type well and an `about.yaml` that says what it is for read well together. ## Ask your assets to say what they are A type can ask each of its assets to carry a description: a file, written by the author or generated, that says what that asset is and what it is for. The people and agents who find an asset read it to decide whether it fits, and the author knows what it is for at the moment of making it. **1. Declare the file with a `described` check.** This is the whole requirement: ```yaml files: required: - path: "README.md" # any file name works kind: "file" content: - { check: "described", message: "Say what this widget is for and where it goes." } ``` With that in place: - An asset whose file is missing, empty, or still has a line starting with `[[describe]]` does not meet the type's requirements. `asset.validate` names it, `zm status` and `world.publishBlockers()` list it, and a push that carries it is refused with `message`. - `services describe` finds it (`mode = "find"` lists it and the price) and can write the description for it (`mode = "all"`). It writes only this file. For a YAML file that holds several texts, name the keys and their limits in bytes, the way this type does for `about.yaml`: ```yaml content: - check: "described" keys: - { key: "intent", max_bytes: 800 } - { key: "usage", max_bytes: 1200 } message: "Write intent and usage." ``` `headings` names sections a Markdown file must have. `assetTypes/README.md` has the full check. **2. Optionally, steer what gets generated with a `describe:` block:** ```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>" ``` `guidance` says what a description of your type covers, and `examples` names assets whose descriptions show the shape you want. Without the block, generation still works from the general describing guide and the check's `message`. An example is an asset of your type in the library whose description is a true, written one, never a scaffold's placeholder, and never the asset being described. When the library has none, leave `examples` empty rather than name one that is not. A type whose assets the model cannot perceive (a sound it cannot hear, a motion it cannot see) says so in `guidance`: the description comes only from what the asset's own files state, such as its name and the `.metadata` where a generation service records the prompt that made it, and when they state nothing the model answers `cannot` with that reason. `services describe` then fails that one asset with the reason and leaves its file as it was. **3. Start the file unwritten in your template.** Put a heading and one line that starts with `[[describe]]` and a short note on what to write, in `template/<file>`, so a new asset starts unwritten rather than with text that only looks written. Do not put example content in it: every new asset carries the template verbatim, and the same text on many assets makes them impossible to tell apart in search. Types whose assets say what they are through their own content (a mesh, a texture, a sound) leave this out: declare nothing and nothing is asked. ## The `behavior.luau` contract `behavior.luau` is the type's code. It returns ONE table, and the keys below are the ones the framework reads off it for EVERY type — the surface you get by declaring them. Anything else on the table is your own: reachable by name through `require("@builtin::assetTypes.assetType.shared.ref").loadTypeModule("<typename>")` for a system that knows your type (`computeShader` publishes `dispatchKey` that way), and never called by the framework. ```luau local M = {} M.ref = { ... } -- methods on every AssetRef of this type M.modules = { ... } -- shared code the instances reach M.global = {} -- reserved key; ship the empty table function M.onCreate(name, opts) ... end return M ``` ### The tables | Key | Shape | What the engine does with it | |-----|-------|------------------------------| | `M.ref` | `{ [name] = function(self, ...) }` | Every function becomes a method on every `AssetRef` of this type: `ref:name(...)`, with `self` the ref. This is how a type gives its instances an API. Two names in here are contracted — see below. | | `M.modules` | `{ [name] = require(".sibling") }` | Shared code the type's instances reach through `asset.containing(__FILE__).modules.<name>`, resolved by the instance's pinned `typeRef` guid. See the section above. | | `M.refShapes` | `{ [method] = "TypeName" }` | Declares the result type of a `M.ref` method so the LSP can check what a call site does with it. See the `M.refShapes` section below. | | `M.events` | an events schema | The events an ASSET of this type fires, declared the way a component declares its own. Readers subscribe through `ref.events.<name>:connect(...)`; firing authority stays with the type. Keyed by the asset's guid, so every resolver of the same asset shares the signals and a re-resolved ref re-attaches to subscriptions already there. | | `M.namePattern` | a Lua pattern string | The name shape `asset.create` enforces for instances of this type, replacing the default `^[A-Za-z][A-Za-z0-9_]*$`. | | `M.global` | `{}` | Reserved. Ship the empty table. | ### The lifecycle hooks Each is optional; a type that omits one costs nothing. The engine calls them: | Hook | Signature | When it fires | |------|-----------|---------------| | `M.onCreate` | `(name, opts) -> { [filename] = contents }` | On `asset.create("<type>", name, opts)`. Returns the files that scaffold the new instance, as a map of relative path to contents; the framework writes them. The `opts` type annotation is the schema `asset.create` validates the caller's arguments against, so annotate it. | | `M.onReplace` | `(dest, files) -> { [filename] = contents }` | On an `asset.create` that writes over an asset already at `dest`, after `onCreate`. Receives the files `onCreate` produced and returns the files to write, so a type that re-derives its content (a rig re-imported from its source file) can carry what an author added to the asset it replaces. Optional. | | `M.onRegister` | `(self)` | Once per instance, the first time it registers: on `engine.onWorldLoaded` for instances already in the world, immediately on a live `asset.create`, and when an instance arrives after that sweep (installed, pulled, or synced from a peer). Guarded by the instance's path, so a double trigger never double-registers. An instance that leaves the runtime releases the guard, so one arriving at the same path again registers. This is what makes "write an asset into the world and it takes effect live" work for content that registers into a runtime registry. The `@builtin` library is excluded from the sweep. | | `M.onChange` | `(ref, change)` | On every VFS write to a file INSIDE one of this type's instances. `change` is `{ path, asset, type, kind, origin }`: `kind` is `"edited"` or `"seeded"`, `origin` is `"local"` or `"remote"` (an importer runs on the originator only). **Filter on `change.path`**: the hook fires for any file under the folder, so act on the one you care about. It runs as its own task and may wait on what it reads: a `vfs.read` of bytes only the blob store holds (a binary file on the web build) fetches them. The writes to one asset reach the hook in the order they were made, each once the hook before it has finished. A re-entrancy guard suppresses writes back into the same asset routed inline from the hook, and a write the hook makes is routed on a later frame like any other, so the hook must be convergent on its own: diff the meaningful state and short-circuit while your own regeneration is in flight. | | `M.onDelete` | `(ref)` | When an instance of this type leaves the runtime: it is deleted, or the engine swaps away from the world binding that held it. Its files are gone from `/zero/source` by then. A removal routed while the instance's `onChange` is waiting reaches `onDelete` once that hook, and every write queued behind it, has finished. On a swap the instance still exists in the world it belongs to, and a VFS write the hook makes reaches the world now bound. | | `M.validate` | `(assetRef) -> { { code, message, severity? } }` | On `asset.validate`, after the structural `type.yaml` check, for the type's own SEMANTIC validation. `severity` defaults to `"error"`; error-severity problems flip `ok` to false, warnings do not. A hook that raises or returns a non-table is itself reported as a `validate.hook_failed` error. `world.push` runs this per user asset, so declaring it enforces your type's rules at publish time with no further wiring. | ### The two contracted `M.ref` names Most `M.ref` methods are your type's own surface: name them what you like, return what you like. Two are read by the engine and mean the same thing for every type, so their shape is fixed: - **`instantiate(self, target?, opts?) -> (root, idMap)`** — makes an instance part of the scene. Defining it is the whole opt-in: `ref:canInstantiate()` is true exactly when it exists. Its **return is a contract** — see the next section. - **`inspect(self) -> detail`** — the type-specific half of `asset.inspect`. See its section below. ### Names a type cannot shadow Some keys resolve on every `AssetRef` before per-type dispatch, so an `M.ref` entry of the same name is never reached: `canInstantiate`, `getSource` / `getBytes` / `getText`, `exists`, `deps`, `meta`, `runtime`, `events`, `modules`, `typeRef`, and the residency flags `has_backing_asset`, `has_runtime_changes`, `cpu_resident`, `gpu_resident`. These are the behaviours every asset must expose identically, which is why they win. ### Reloading A `behavior.luau` edit ripples to existing refs with no restart: dispatch reads `mod.ref` fresh each time and `require`'s cache is re-run in place. A type whose folder appears at runtime is picked up on the next dispatch — negative results are not cached either. ## The names your type's assets hold (`references:`) An asset that names another asset by a string depends on it: the named asset travels with it into a pack or a pull, and a name nothing answers to is an unresolved dependency `worldValidation` reports and the push gate refuses. The engine reads `require(...)` and the asset lookups on its own; a name your type's format or API carries anywhere else is one you declare, in `type.yaml`: ```yaml references: # A top-level key of this type's own files that holds a name (or a table of them). keys: - { key: "shader", type: "shader" } # A call's argument, counting from 1, wherever the call is written. args: - { call: "g:node", position: 2, type: "procNode" } # A field of the options table a call takes. fields: - { call: "fx.beam", field: "material", type: "material" } ``` `type` is the category the name resolves in, one of `asset.categories()`. An argument declared once per category, at the same `call` and `position`, takes the name from whichever category answers. A key or a field holds one category: declared again with another `type`, it takes the later one. The block is declared as the type registers, when the library or the world holding it is laid into the VFS, so every reference table reads it whatever script has or has not run. `asset.declareReferenceKey`, `asset.declareReferenceArg` and `asset.declareReferenceField` declare the same things from a script, for a surface that is no type's. ## What search does with your type (`indexing:`) A type declares what search embeds for its instances. **This is not optional and there is no useful default**: ZeroMind runs your declaration and adds nothing of its own, so anything you do not point at is absent from the index — silently, and for every instance of the type forever. Twenty types once shipped with no `indexing:` block and every instance of them was unfindable. A type whose instances are never searchable (a run record, a log) says so with `indexing: none`; every other type declares a recipe. A type without an `about.yaml` beside its `type.yaml` refuses every push that carries it, whether or not the push edited the type. Texts in it that are not written yet follow the rule every `described` file follows: they refuse the push that edits, stages or commits the type. Two words carry the whole thing, and they are the same two the search tool exposes: - **`identity`** — what an instance IS. For anything you can look at, that is the picture (`content:` + `modality: image`, matched directly by a text query). For everything else it is the authored text that says what it is. - **`capability`** — what it DOES and how it is made. Code, settings, the model-written summary of them. A type with no behaviour has no `capability` entry. A type with nothing to look at has no image. Leaving a slot empty is a statement; filling it with whatever happens to be lying around is not. The one you have to decide when authoring a type is: *for an instance of this, what is the thing a person would recognise it by, and what is the thing it does?* A texture answers "the image" and "its compression settings". A module answers "its README" and "its code". Write those two answers into `indexing:` and the rest follows. Three specifics worth knowing before you write one: - **`derive` is instructed by this type's own `README.md`.** A model reads an instance's source and writes a description; what it embeds is the description, not the source. It follows your README to know what it is looking at, so a type whose README says what its instances are gets good derivations for free. There is no prompt to name. - **`derive` loses detail.** It keeps the main technique and drops secondary ones. If the source is code you want searchable by its own vocabulary, declare the same files a second time as `capability` with `extractor: verbatim`. Two entries, same role, different failure modes. - **Never `source: { field: name }`.** A filename is one or two words with no usable embedding, and a corpus indexed that way answers "language runtime" with `say_runtime_2.soundClip`. `facets:` is the second stage: a `file:` embedded whole (`.metadata`, so keys nobody declared still get indexed) plus computed keys that state a value for every instance including `false` — the raw file can imply "not rigged" and can never say it. The full schema, every extractor and the facet sources are in [`assetTypes/README.md`](../README.md); the scaffolded block with inline guidance is in [`template/type.yaml`](template/type.yaml). ## A container whose contents work together (`system: true`) A type that is a container can say its contents are one working whole by declaring `system: true` at the top level of its `type.yaml`. `package` declares it: the components, modules, scenes and tools a package ships are parts of one system. The declaration is plain data, read from the decoded `type.yaml` (`require("@builtin::assetTypes.assetType.shared.declaration").of("package").system`), and a type that leaves it out is an ordinary type. ## What your type looks like (`preview:`) A type declares how its instances are pictured, and the engine makes the picture when something needs it: staging an instance whose `preview.png` is missing or pictures content it no longer holds queues a render, a view that shows the instance asks for one, and `ref:preview()` renders one on demand. An edit renders nothing on its own. Each `preview.png` records the content it pictured, which is how a stale one is told from a current one. The preview studio owns the light, the backdrop, the framing and the camera, and nothing in the world reaches it, so an instance pictures the same in every world. Declare one of four recipes: | `kind` | The picture is | Your type provides | |---|---|---| | `instantiate` | the asset stood in the studio through its own `instantiate` | `ref.instantiate` | | `image` | a file the instance holds, fitted to the frame, nothing rendered | `source.files`, the candidate files in order | | `surface` | the asset painted on a builtin mesh | `mesh`, and `ref.previewMaterial(self) -> { material }` | | `custom` | whatever `ref.preview(self, opts)` returns | `ref.preview` | ```yaml preview: kind: surface mesh: sphere ``` A scene-instantiable type that declares nothing still answers `ref:preview()` through its `instantiate`, but persists no picture: the declaration is what makes the engine keep one. List `preview.png` among the type's files so an instance may hold it. The picture is findable only when `indexing.content` names it. The two are separate statements, and neither is derived from the other: ```yaml indexing: content: - role: identity modality: image source: { file: "preview.png" } ``` `@builtin::assetTypes.assetType.shared.previewRecipe` reads the declaration; the scaffolded block with inline guidance is in [`template/type.yaml`](template/type.yaml). ## Typing a result from the instance (`M.refShapes`) Every method on `M.ref` is checked against the type its `--!return` names, and every asset of the type gets the same one. That is right for most methods and wrong for the ones whose result is shaped by the ASSET: a method answering one entry per child folder, per row of a config, per binding an input map declares. The widest true annotation for those is `{ [string]: Thing }`, and an indexer accepts every key — so a caller's typo reads as valid and nothing reports. `M.refShapes` lets the type state the result per instance. Each entry is `function(self) -> (typeExpression, source?)`: a Luau type expression for THIS asset, and the module whose type vocabulary the expression uses. Read the asset; never run it — stating a type must not take effect. ```lua -- inputMap.assetType/behavior.luau — the shipped example. -- `activate()` hands back one handle per control the map declares. 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)` resolves and `map.noexisting` is reported with the map's real controls. Without it, both pass silently. ### What the call site has to say A per-instance type is matched by the asset's IDENTITY, so it applies only where the reference the method is called on says WHICH asset. Two spellings do: ```lua -- a component field, whose declared default names the asset public = { map = Field.assetRef("inputMap", "@builtin::inputMaps.default", Sync) } local controls = public.map:activate() -- typed for THAT map -- an annotation, naming category and identity local m: AssetRef<"inputMap", "@builtin::inputMaps.default"> ``` `asset.resolve("<identity>", "<category>")` carries the CATEGORY, so the methods the category defines are checked on the result — but not which asset it found, so a per-instance result keeps the category's declared return. A reference that names no asset (`asset.resolve(someVariable)`, a value passed in as a parameter) carries neither, and calls on it are unchecked. That is the same rule everywhere: the checker states what the code states, and a name computed at runtime states nothing. ### The result is a value, not a binding The type belongs to the expression, so it travels the way any inferred type travels. Indexing the call's result is checked; stashing it in a module local and indexing it from another function is not, because the local's declared type is what carries across — and `any` (or no annotation) carries nothing: ```lua -- checked: `d.greting` is reported function awake() local d = public.dialogue:lines() d.greeting:say() end -- NOT checked: `lines` is a module local typed `any` local lines: any = nil function awake() lines = public.dialogue:lines() end function start() lines.greting:say() end ``` Bind what you need at the call site and the names stay checked in both places — the asset-derived type at the boundary, ordinary scoping after it: ```lua local lineGreeting = nil function awake() local d = public.dialogue:lines() lineGreeting = d.greeting -- `d.greting` is reported here end function start() lineGreeting:say() end ``` ### When nothing is reported and you expected something From a call site, a type that is correct and a type that was never published look the same: no diagnostic either way. `require("@builtin::assetTypes.assetType.shared.refShapes").published()` answers which it is. It returns two maps — the category surfaces, and the per-asset results keyed by identity and then by method: ```lua local shapes = require("@builtin::assetTypes.assetType.shared.refShapes").published() shapes.categories.dialogue --> "{ lines: () -> any, lineNames: () -> { string } }" shapes.returns.Shopkeeper.lines --> "{ browsing: Line, farewell: Line, greeting: Line, wares: Line }" ``` An identity absent from `returns` was never published for, which is a different thing from published-and-correct — and the reason to look here rather than at the call site. The answers are recomputed when an instance is written, so a control added to a map reaches the checker with no restart. Only types that declare `refShapes` are read per instance, and at most 64 assets of one type are — past that the type keeps its declared return and the engine log names what was dropped. ## The instantiate hook (`M.ref.instantiate`) — a contracted return A type opts into becoming part of the scene by defining `instantiate` on its `ref` table. That is the whole opt-in: `ref:canInstantiate()` is true exactly when the hook exists, so consumers offer a scene path — an `Asset.source` field, a viewport drop, a tool argument — by CAPABILITY rather than by a list of type names. Unlike `inspect`, this hook's **return is a contract**. A caller writes one piece of code against every instantiable type, so what comes back cannot vary by type: ```lua -- behavior.luau local Instantiable = require("@builtin::assetTypes.assetType.shared.instantiable") local M = {} M.ref = { instantiate = function(self, target, opts) local root = Instantiable.root(self, target, opts) -- IN: the base opts -- ... compose whatever this type is, under `root`, NOW ... return Instantiable.result(self, root, idMap) -- OUT: the contract end, } return M ``` **IN.** `target` is an owning entity ref: the instance lands under (or, for a hierarchy type, onto) that owner. With no target the type spawns a fresh root. `position`, `rotation`, `scale`, `name` and `temporary` mean the same for every type — `Instantiable.root` applies them to a root you mint, `Instantiable.place` to one you adopt, so you never re-read the spec. Honour more opts of your own if your type needs them, and document them in the hook's `--!arg` docs. **OUT.** Return `(root, idMap)` through `Instantiable.result`: - **`root`** — the composed root, as an `EntityRef`. Composition is **synchronous**: everything your type builds is live when you return, so the caller can parent to it and read its components in the same statement. Do not defer the work to a component's `awake` and return an empty shell — a root that is not live is refused. - **`idMap`** — the `originalId -> runtimeId` map naming what you spawned, or nil for a type with no addressable children (`result` normalises it to `{}`). **A caller never receives nil.** A component that re-composes your asset on every load keeps this map and passes it back in, which is how a cross-entity reference into the composition survives a reload. `AssetRef` runs every `instantiate` through `result` on the way out whether or not your type called it, so returning something else fails at your own call rather than handing a caller a nil root. Calling it yourself is still how you say what you return, and running it twice changes nothing. `@builtin::assetTypes.assetType.shared.instantiable`'s README is the full reference for both halves. ## The inspect hook (`M.ref.inspect`) `asset.inspect(ref)` resolves `ref`, builds a common envelope shared by every asset (`identity`, `name`, `guid`, `source`, `typeName`, `typeDefinitionPath`, `scope`, `origin`, `description`, `tags`), then calls the resolved type's own inspect hook — `M.ref.inspect(self)` on `behavior.luau` — for the type-specific `detail`. The hook returns *only* `detail`; the framework fills in everything else: ```lua -- behavior.luau local M = {} M.ref = { inspect = function(self) -- self.path is the asset's own folder. Parse the asset's OWN -- files — source text, a native payload header, its `.metadata` -- — never runtime state (a live component instance, a GPU -- upload, a compiled schema): inspect must work on an asset that -- hasn't been compiled, uploaded, or registered yet. return { -- type-specific fields, whatever this type wants to surface } end, } return M ``` A type that omits `M.ref.inspect` yields the generic record: the envelope alone, `detail = nil`. That's a valid, working state — not an error — but it means agents calling `asset.inspect` / `tools.use("assets","describe", ...)` on an instance of the type see only identity and scope, none of what makes an instance of it meaningful. Declaring the hook is what makes a type's instances inspectable for what they actually are. A hook that raises leaves `detail = nil` and sets `record.warning` to the error text — `asset.inspect` itself never raises for a resolved ref. ## Discovery - `asset.list("assetType")` — every registered asset type. - `asset.inspect("<typename>")` — the type's identity, source path, and this README, plus its `about.yaml` texts as `detail.about`. - `asset.resolve("<typename>", "assetType")` — the `<typename>.assetType` asset itself (used by the validator to find the type's `type.yaml`). ## Related - `assetTypes/README.md` — the full `type.yaml` schema and validator rules. - `@builtin::assetTypes.assetType.shared.instantiable` — the instantiation contract's shared implementation, both halves. - Every other `*.assetType/` here — the concrete types built on this meta-type. ## The Inspector a type ships Every type names the file its Inspector sections come from with `inspector: { type: "inspector.luau" }` in its type.yaml. `asset.create("assetType", name)` scaffolds the block and an `inspector.luau` whose Contents section lists an asset's files, built from `ctx.parts`. `asset.validate` on a type that names none reports `assetType.inspector_required` (an error for a `@builtin` type the engine's library authors; a warning for a world's own, and for one the library carries by reference, which the world that published it answers for), and one whose file does not load `assetType.inspector_invalid`. ## In the Inspector Type: what the type.yaml declares (suffix, container, primary files, the settings and inspector blocks), whether the type ships a behavior, how many instances it has, and **Open** for its files. Schema: the settings every instance carries, as read-only typed controls at their defaults (a type whose schema is per instance says where it is declared). Instances: a link to each of the first 20 assets of the type, and how many there are when there are more.

# Tool (asset type) A tool is a single agent-callable function whose schema is its `typed function` signature. Tools are the workflow surface — composed Luau operations that bundle multiple engine APIs into one named call, suitable for agents (and for use from `execute`). Each tool has a declared input shape, return shape, and a strict result envelope (`ZmToolResult`). **Every exported tool function is declared `typed function`.** The `typed` keyword is what makes the signature real: it compiles to a runtime dispatch wrapper that checks each argument against its declared type and raises on a mismatch. A plain `function` — even one carrying Luau type annotations — is checked by nothing at call time, so the tool accepts any input and reports success on garbage. See "The typed signature is the schema" below. ## When to use one - You're composing several engine APIs into a workflow that's worth naming and discovering (`tools.list`, `search_tools`). - You want an agent to be able to call this function by name without reading source code — the typed signature is the API contract. - The operation has clear success / failure modes that fit `{ exitCode, data?, stdout?, message? }`. If the function is just a Luau utility called from other code, use a `.module`. If it's stateful background work, use a `.service`. If it's pure GPU compute, use a compute `.shader`. ## Where it lives Tools nest under a `.toolbox/` parent — toolboxes are the namespace. - Source: `/zero/source/.../<Toolbox>.toolbox/<name>.tool/` - Identity: `<Toolbox>.<name>` (dotted form — the toolbox name is part of the tool identity). - Folder shape: - `init.luau` (or `init.lua`) — exports the function, and carries its schema in `--!desc` / `--!arg` / `--!return` / `--!example` annotations above it. The typed signature IS the tool's signature. **Required.** - `README.md` — instance documentation. **Required.** - `.metadata` — tool-level tier and tags. **Required.** ## How to create one Use the workflow tool that scaffolds the right `.tool/` shape under an existing toolbox: ```luau -- The `tools` global authors a tool inside an existing toolbox: tools.create({ name = "<name>", toolbox = "<Toolbox>", description = "What this tool does, agent-facing.", code = "return shared.ok(true)", -- the function body }) -- Or scaffold the folder manually: local box = asset.resolve("<Toolbox>.toolbox") asset.create("tool", "<name>", { into = box }) ``` Always nest under an existing toolbox — if you don't have one yet, call `tools.createToolbox({ name = "<Name>", description = "..." })` first. ## How it operates 1. **Registration.** Writing `init.luau` into a `.tool/` folder registers the tool with the tools registry. It becomes callable as `tools.use("<toolbox>", "<name>", ...)` from Luau, `use_tool { toolbox, tool, args }` over MCP, and `zero <toolbox> <name> [args]` from the engine shell. 2. **Schema validation.** The `typed function` signature declares the parameter shape. Its runtime dispatch wrapper checks each argument before the body runs and raises a message naming the argument, the expected type, and what arrived. 3. **Return envelope.** Tools MUST return `ZmToolResult`: ``` { exitCode = 0|n, data? = any, stdout? = string, message? = string } ``` `exitCode = 0` means success; non-zero means failure, with `message` carrying the human-readable reason. 4. **Discovery.** The tool surfaces in `tools.list()`, `search_tools { query: "..." }`, and `asset.inspect`'s `exposes:` section (parsed from the typed signature and the `--!` annotations). 5. **Hot reload.** Editing `init.luau` re-registers the tool. Subsequent calls see the new behaviour / schema. ## Discovery - `tools.list()` — every registered tool (any toolbox). - `search_tools { query: "..." }` — MCP-side tool search. - `asset.list("tool")` — same set, asset-style listing. - `asset.inspect("<toolbox>.<name>")` — signature, parameters, return shape, this type README. - `ls /zero/source/<Toolbox>.toolbox/<name>.tool` lists the files the folder holds. ## Authoring conventions - **Implementation lives in `init.luau`.** The tool's full function body — argument validation, the call into engine APIs, the return value — goes in `init.luau`. The toolbox's `shared.module/` is for genuinely cross-tool helpers (`ok` / `fail` constructors, shared parsers / validators), not per-tool logic. Do NOT make `init.luau` a one-line forwarder onto a function defined in `shared.module/`. - Use `shared.module/` in the parent toolbox for `ok` / `fail` result constructors. Don't hand-roll the envelope per tool. Pull it in with `local shared = require(".shared")` from any tool that needs the helpers. - Annotate every public function with `--!desc` / `--!arg` / `--!return` / `--!example` directly above the function — the LSP and toolbox surfaces extract these. - **Declare every exported function `typed function M.name(arg: T, opt: U?): R`.** This is not conditional on the signature being "clear" — a loose argument is the case that needs the checked contract most. Give the argument the type it actually accepts (a union, a named table type, a literal enum) instead of dropping to an unchecked `function`. - Keep tools focused. One tool, one named workflow. Split rather than grow. - Tag tools in the sibling `.metadata` (`{ "tags": [...], "tier": "core|content|specialized" }`) so `search_tools` can find them by domain (e.g. `entity`, `scene`, `assets`, `physics`). ## The typed signature is the schema The `typed` keyword compiles the function into a dispatch wrapper that validates arguments at call time. Without it there is no check at all, so the tool runs its body on whatever arrived and returns a success envelope built from garbage. Type each argument as narrowly as the operation actually accepts: ```luau export type Mode = "edit" | "play" export type SpawnOpts = { position: { number }?, parent: EntityRef?, count: number?, } --!desc Spawn a primitive and return the entities it created. --!arg kind Primitive shape to spawn. --!arg opts Placement and repetition options. --!return The spawned entity refs. --!example "cube", { count = 4 } typed function M.spawn(kind: ("cube" | "sphere" | "plane"), opts: SpawnOpts?): { EntityRef } ``` - **A fixed set of accepted words is a literal union**, never `string`. `("cube" | "sphere")` rejects `"Cube"` at the call and lists the accepted values in the error; `kind: string` accepts `"banana"` and fails somewhere deeper, or not at all. - **A structured argument is a named `export type`**, never `{ [string]: any }`. Naming it puts every field and its type on the tool's page and checks them on entry. - **A named type is only as strong as its fields.** `opts: DrawOpts` where `DrawOpts = { value: any }` passes for a table and checks nothing inside it — the same hole, one level down. The argument checker walks nested fields, so type each one; a type name is not a place to put the work off. - **A dispatch table keyed by an `op` string is several tools.** `{ op: string, [string]: any }` is one unchecked entry point wearing N signatures — split it so each operation carries its own checked arguments. - **An entity or asset argument is `EntityRef` / `AssetRef<kind>`**, never `any` and never a bare name string. - `any` is correct only for a value the tool genuinely does not constrain — a passthrough payload, arbitrary user data. It is not a placeholder for a type not yet written down. ## Common pitfalls - **A plain `function` export.** The tool registers and dispatches, so nothing looks wrong — it just never rejects bad input, and every call reports success. Declare it `typed function`. - **Annotations without `typed`.** `function M.f(x: string)` reads as typed and checks nothing at runtime; the annotation only informs the LSP. The keyword is what enforces it. - **Wrong return envelope.** Returning a raw value instead of `{ exitCode = 0, ... }` will be flagged by callers expecting `ZmToolResult`. - **Tool identity.** The fully-qualified identity is `<toolbox>.<name>`, NOT just `<name>`. Toolbox-less tools don't exist by design. - **Missing dependency.** Tools requiring another package / service must declare it; otherwise the call fails at runtime with a missing-module error. ## Related types - `.toolbox` — the required parent namespace for tools. - `.module` — for Luau libraries with no agent-facing schema. - `.service` — for stateful background work, not callable request/response operations. ## In the Inspector Tool: how the tool is called (its signature), what it does, each argument, what it returns, and its examples, with **Open** for its source.

# scene Work with the live scene across its whole lifecycle: start one from a template, understand its structure, reshape its player, clean it, and set it as the world's entry point. ## The player model A scene is either **playerless** (`"none"`) or **spawns a player per joining user** (`"spawns"`). The player is defined by a **PlayerPrototype** subtree — a body (the avatar a joining user becomes) and a camera rig. Zero is like Roblox here: the default character is a socket a joining player's own avatar replaces, so **reuse the canonical player setup rather than hand-rolling one** — a divergent player breaks multiplayer. `scene.summary` and `scene.tree` always report the scene's player intent, so you can see a scene's type without knowing this tool exists. ## Start a scene - **create** — create a NEW scene from a typed template (`empty`, `flat`, `player`, `menu`) and load it. Omit the template to get the catalog of what each contains. `opts.startup = true` also makes it the world's entry scene. The starting point is a choice, not a scene handed to you silently. - **setStartup** — make the currently active scene the world's startup scene, so it survives world reloads. No argument, no clear. ## Understand a scene - **tree** — the entity hierarchy as an indented tree, led by a one-line player-state header. The quick "what is the current structure" view. - **find** — the single entity-discovery tool over the whole live population. Search by `name`, `component`, or `attribute` (each a list), scope by hierarchy or scene, and add projections with `include`: `transform` (`position`, `rotation`, `eulerAngles`, `scale` as arrays, read `position[1]`), `componentValues` (each component's live field values), `attributes`, `scriptComponents`. Every match carries `components` (the component names) without one. Returns `{ matches, total, truncated, searchedAxes }`. - **summary**: entity + root counts, the top-level rows (at most `limit`, 20 when omitted), the active camera, the **player** setup (intent + counts + the verb to change it), and the world's **startup** scene. `under` scopes the counts and rows to the subtrees of named entities, and `fields` keeps only the named fields, so `{ fields = { "entityCount" } }` answers the count alone. - **list** — the SCENES you can load, so you find one by name instead of guessing. `scope` filters `world` (default, this world's own scenes), `library` (builtin `@builtin::` templates / demos / canonical setups), or `all`; each entry says whether it is currently loaded / active. - **loadedLayers** — the scene LAYERS composing the live scene right now: the root scene plus any additive overlays (editor UI, effect passes), each with its backing scene, status, visibility, and additive flag. A layer whose load produced failures reads `status = "partial"`. ## See what a load did, and what a scene costs - **observe** — what every load did and what each loaded scene costs. `lastLoad` is the last load's report: what it produced, the root it replaced and the overlays that went with it, the entity counts on each side, the milliseconds each phase took, and every failure with the entity, component or lifecycle hook it is about. `layers` counts what the engine attributes to each layer, `cost` is each layer's per-frame entrypoint tick summed across the window `window` reports. - **whyPartial** — why a layer is not whole: the nearest cause from a closed set, the engine's own words for it, and the full failure list. The answer to "the scene loaded but something in it is missing". - **cost** — which loaded scene is eating the frame time, ranked by what each layer's entrypoint tick costs. A scene entrypoint's per-frame `update` / `editorUpdate` runs on the scheduler rather than as a component, so `profiler.scripts` does not list it; this does. Pass `true` to open a fresh window, let frames pass, then call again. ## Reshape the player - **player** — read the setup, flip the typed intent between `"spawns"` and `"none"`, and reshape: `body` adopts an entity or an avatar asset ref, `camera` wires an entity carrying a Camera, `behavior` sets the player camera's behavior to any `cameraBehavior` component. `"none"` removes the prototype and spawns; `"spawns"` authors the canonical setup when missing. - **spawns** — the spawn POINTS: where joining players arrive, as opposed to `player`, which decides whether they arrive at all and what they become. No argument reports every PlayerSpawn with its transform, the prototype it instantiates (and whether that reference still resolves), and its team / role / maxPlayers / spawnPolicy / placement. `"add"` places one, `"set"` reconfigures one (omitted fields keep their value), `"remove"` deletes one; `opts.spawn` picks WHICH when the scene has several. A spawn's transform is where a body lands and its forward axis is the heading it faces — the edit-mode overlay draws that ring, arrow and standing capsule. ## Change the content - **clean** — clear the scene down to a protected keep-set. By default only the player setup survives (never orphaned); `keep` protects a `scene.find` selection (e.g. `{ component = { "Light" } }`); `keepPlayer = false` wipes the player too; `dryRun` previews. Only the active scene layer is touched. - **remove** — delete entities by name, id, array, or a `scene.find` query, subtree-deep. Removing the default floor is `scene.remove("ground")`. - **save** — save a loaded scene back to its own `scene.json`; a no-arg call reuses the last scene name. - **load** — load a scene snapshot into the live world. - **reload** — re-run a loaded scene's build after editing its entrypoint. `load` of an already-loaded, unchanged scene is an idempotent no-op; this rebuilds it, in edit mode. Omit the name to reload the active scene. A rebuild saves the scene, as a write to `build.luau` does, so `scene.json` holds what the reload built and the next play loads it. A library scene (under `/libs/`) is rebuilt in the live scene alone. - **replace** — replace a loaded scene's content with another scene's content, keeping the target's identity. Supports `dryRun`. ## Maintenance - **migrate_v6_to_v7** — convert a v6 scene's `scene.json` to the v7 format in place. User-initiated only. World mode (play/edit) and additive layers are `wld`; accepting or rejecting play-mode changes is `sceneAuthoring`.

# action Perform a gameplay action by the name of the control the game reads it from — tap `jump`, walk with `move`, hold `sprint` — with no device in between.

# bindings Every control that is live right now, and which map contributed it.

# click A full mouse click, optionally moving the cursor first.

# device Read the active device class, or force one so a desktop session can be checked as a phone.

# drag Drag a finger across the look zone.

# fired Which controls fired since the last call — the assertion half of a simulated input.

# held What the session is holding right now — keys, mouse buttons, fingers, the on-screen stick, and connected pads. `release` lets go of exactly this set.

# key Tap a key — down, optional hold, then up.

# keyDown Press a key and leave it held until `keyUp`.

# keyUp Release a key held by `keyDown`.

# layout What the on-screen touch controls are drawing right now.

# sim.listMacros List every registered input-macro asset in the project, returning an array of asset handles ready to hand straight to `sim.macro` for playback. Use it to discover which recorded or authored macros are available.

# lockPointer Lock or release the pointer.

# inputSim.macro Play back a macro of timed input events, driving the keyboard, the mouse, a gamepad and the glass in one sequence. Accepts an inline event list, an input-macro asset handle, or the identity or path of a saved input-macro asset; playback runs as a background task that waits between events according to each event's timestamp. Each event is `{ t, op, arg }`, where `t` is seconds since the macro started (events play in order of `t`, and events sharing one `t` play in the order the list writes them), `op` names one of this toolbox's own verbs that drives input, and `arg` carries that verb's arguments: the value itself when the verb takes one, a table keyed by the verb's parameter names when it takes several. So the keyboard and mouse verbs (`keyDown`, `keyUp`, `key`, `mouseDown`, `mouseUp`, `mouseMove`, `mouseMoveBy`, `click`, `scroll`, `lockPointer`), the pad verbs (`pad`, `padOff`, `padDown`, `padUp`, `padPress`, `padStick`, `padTrigger`) and the touch verbs (`touch`, `touchDown`, `touchMove`, `touchUp`, `stick`, `tapButton`, `drag`, `pinch`) all play, and a point written for a verb is the point the op of that name takes. A `touchDown` event's `id` is the label its contact goes under for the run of the macro, and the `touchMove` and `touchUp` events carrying that label drive it. The same primitive drives a player, a car, or a UI test; author macros by hand or record them with `inputSim.recordMacro` and persist them with `inputSim.saveMacro`.

# mouseDown Press a mouse button and leave it held.

# mouseMove Move the cursor to an absolute screen position.

# mouseMoveBy Move the pointer by a relative motion.

# mouseUp Release a mouse button held by `mouseDown`.

# pad Connect a simulated gamepad and return its slot.

# padDown Hold a canonical pad button down.

# padOff Disconnect a simulated gamepad.

# padPress Press a canonical pad button, named by position.

# padStick Push a pad thumbstick to a deflection.

# padTrigger Squeeze a pad trigger to an analog travel.

# padUp Release a held pad button.

# pinch Pinch or spread two fingers.

# inputSim.recordMacro Record live input over a window of time on all three device classes, emitting a timed event whenever something changes. It reads what every device is holding each frame (keys, mouse buttons, pad buttons, the pad itself, fingers on the glass) and what each one is measuring every `sampleInterval` seconds (mouse motion, the wheel, thumbsticks, triggers, a sliding contact). It returns the captured events array plus the task handle so the caller can cancel early, and stops on its own after `duration` seconds elapse or when the optional `until_` predicate answers true. It takes those arguments the way every verb in this toolbox does, positionally, how long first, so `recordMacro(3.0)` records for three seconds and `recordMacro(3.0, 0.02)` samples that window more finely. The result works for any input-driven scenario, a player walk, a car drive, a pad rhythm or a UI flow, and can be replayed immediately with `inputSim.macro` or persisted with `inputSim.saveMacro`.

# release Put everything down between takes: fingers, the stick, held keys and mouse buttons, and connected pads.

# sim.saveMacro Persist a recorded or hand-authored event list as a named input-macro asset in the world's source, returning the asset handle of the saved macro for immediate playback. The name becomes the asset's folder name and must be non-empty. Use it to turn a `sim.recordMacro` capture into a reusable, replayable macro.

# scroll Turn the mouse wheel.

# inputSim.shared The point conversion the toolbox's absolute-point pointer verbs share. `click` and `mouseMove` take a screen point in physical pixels, in the space the input's receiver reads positions in. The input primitives beneath them place a pointer on the window. `windowPoint(x, y, target)` carries a point into that window space. For the game (`target` `"game"` or nil) the point is in the space `capture` reports, the scene's own pixels measured from the top-left of the scene's rect, and the conversion adds that rect's origin; where the scene is drawn across the whole window (a headless engine, a fullscreen play session) the rect starts at the window's top-left and the conversion is the identity. For the UI (`target` `"ui"`) the point is already on the window, where a widget sits at its `gui.bounds` rect times `ui.pixelRatio()`, and passes through unchanged.

# stick Push the on-screen movement stick to a deflection.

# tapButton Tap an on-screen touch button by the label a player sees.

# touch A whole finger tap: contact, optional hold, lift.

# touchDown Begin a finger contact and leave it down.

# touchMove Move a finger that is already down.

# touchUp Lift a finger.

# Toolbox (asset type) A toolbox is a namespace folder that groups related `.tool/` children plus shared helpers. Toolboxes are the addressing layer for tools — every tool's identity is `<toolbox>.<name>`, so the toolbox name appears in every callsite. Toolboxes also own a `shared.module/` helper module that the toolbox's tools `require(".shared")` for common utilities (result envelopes, validators, formatters). ## When to use one - You have two or more related tools that belong together (an `anim` toolbox with `play` / `stop` / `list`; an `assets` toolbox with `compose` / `validate`). - You need a shared helper module exclusively used by a tool set. - You want a clean dot-namespaced identity for your tools (`my_tools.foo`, `my_tools.bar`). If you only have one tool and no shared helpers, you still need a toolbox — it's the required parent. ## Where it lives - Source: `/zero/source/.../<Name>.toolbox/` - Identity: `<Name>` (the `.toolbox` suffix strips). - Folder shape (container-style — open child set): - `README.md` — instance documentation. **Required.** - `.metadata` — agent-editable tags + free-form fields. **Required.** - `shared.module/` — optional shared helper module (`init.luau` + `README.md`). - `<x>.tool/` — any number of `.tool/` children. The validator permits unlisted children (`allow_unlisted: true`) so tool subfolders don't need to be declared per-name. ## How to create one ```luau tools.createToolbox({ name = "<Name>", description = "..." }) -- or the generic scaffold: asset.create("toolbox", "<Name>") -- Creates: /zero/source/<Name>.toolbox/ -- README.md (instance README template) -- shared.module/ (ok / fail result-envelope constructors) -- `folder` places it in a subfolder of /source; `into` nests it in a -- resolved `.package`: asset.create("toolbox", "<Name>", { folder = "tools" }) ``` Then add tools under it: ```luau local box = asset.resolve("<Name>.toolbox") asset.create("tool", "<toolName>", { into = box }) ``` ## How it operates 1. **Registration.** Writing into a `.toolbox/` folder indexes the toolbox. It becomes the parent namespace for any `.tool/` children. 2. **Tool identity composition.** Each child `.tool/`'s identity is `<toolbox>.<tool>`. Listing the toolbox's tools enumerates the children. 3. **Shared helpers.** Tools inside the toolbox pull `shared.module/` in via `require(".shared")` (relative require from the tool's `init.luau`). 4. **Hot reload.** Editing `shared.module/init.luau` reloads every tool in the toolbox (they all `require` it). Editing the toolbox folder structure (adding / removing tools) reflects in `tools.list` immediately. ## Discovery - `asset.list("toolbox")` — every registered toolbox. - `asset.inspect("<name>")` — child tools, source, this type README. - `ls /zero/source/<Name>.toolbox` lists the files the folder holds. - `tools.list()` — every tool across every toolbox. ## Authoring conventions - **Per-tool implementation lives in each `<x>.tool/init.luau`.** The tool's full function body — argument validation, the call into engine APIs, the return value — goes in `init.luau` with `--!desc` / `--!arg` / `--!return` / `--!example` annotations above the function. **Every exported function is declared `typed function`** — the keyword compiles to the runtime argument check, so a plain `function` accepts any input and reports success on garbage. Do NOT make `init.luau` a one-line forwarder onto a function defined in `shared.module/`. - Keep `shared.module/` thin: result envelope constructors (`shared.ok(value, formatted)`, `shared.fail(message, exitCode)`), argument validators, common formatters — anything genuinely used by more than one tool. Per-tool logic does NOT belong here. - Group by domain, not by surface — `assets` toolbox holds asset operations, `entity` toolbox holds entity operations. Tools that feel unrelated probably want different toolboxes. - Document the toolbox's surface in `README.md` — what problem space it covers, who calls into it, any shared invariants. ## Common pitfalls - **Missing parent toolbox.** Tools cannot exist outside a toolbox. Always scaffold the toolbox first. - **Cross-toolbox shared.** Each toolbox has its own `shared.module/`. Don't try to share helpers across toolboxes; if you need cross-toolbox helpers, factor them into a standalone `.module`. - **Toolbox identity.** Strip the `.toolbox` suffix in identity references (`my_tools`, not `my_tools.toolbox`). ## Related types - `.tool` — the required child(ren) of a toolbox. - `.module` — for cross-toolbox utilities. ## In the Inspector Tools: how many tools the box holds, whether it ships a shared module, what each tool does, and a link to each tool.

# camera Author and manage the cameras in a scene: create one, list every camera, read its full parameters, change them, aim it, frame a subject, or make it the primary camera. A physical camera on the machine (a webcam, a phone camera) is `cameraInput`, covered by the `topics/camera-input` guide. ## Create & inspect - **create** — spawn a camera entity at a position, with any starting parameters in one call: fov, near/far, priority, render layers, a follow target, a behavior, a `lookAt`, and `target` (render-to-texture — `true` or `{ width, height, name }` renders the camera offscreen instead of to the viewport). Returns the new camera's full parameter record. - **list** — every camera in the scene, enumerated from the live ECS so a just-spawned, hidden, or runtime camera is never missed, sorted by descending priority, each row marking whether it is the on-screen (`active`) camera, whether it is the `player` camera (the one the PlayerPrototype uses), and its main/editor role. - **get** — the full parameter record for one camera (position, rotation, fov, near, far, priority, render layers, debug channel, render target, behavior, follow). Defaults to the active/main camera when you name none. ## Modify - **set** — change any of these camera parameters in one call: `fov`, `near`, `far`, `priority`, `debugPass` (the diagnostic view this camera renders, by name), `debugChannel` (the same as a raw numeric channel), `renderLayers` (the layer-name spec, e.g. `"all !ui"`), `behavior` (a `cameraBehavior`-tagged component), and `follow` (the entity the behavior tracks). Pass no camera to modify the active one. A camera's render target is set when it is created (`create` with `opts.target`). - **lookAt** — aim a camera at a world position or at another entity. - **frame** — position and aim a camera to frame a target entity from a distance and angle in one step. - **bindToPlayer** — make the joining player use a camera in play by binding it to the scene's PlayerPrototype: it re-parents the camera into the prototype (so it is cloned with the player) and sets the prototype's `camera` role to it. Pair it with a static behavior (`set { behavior = "@builtin::controller.menu" }`) for a fixed PS1-style camera the player moves within, or a follow behavior (`orbital_follow`, `third_person_follow`, `first_person`, …) for a tracking one. A player-less scene has no prototype — there the rendering camera is simply the highest-priority one, ordered with `set { priority = N }`. ## Related surfaces - Capturing an image from a specific camera is the `capture` toolbox (`capture` with a camera, or `oneshot`) — it renders any camera's authored pose offscreen without changing what is on screen. - Putting geometry ON a render layer is the `renderLayer` toolbox (`renderLayer.set`); `renderLayer.camera` is the layer-system-side way to set a camera's include filter, the same field `camera.set { renderLayers = ... }` writes. - The scene's primary camera is normally spawned for you from the scene's PlayerPrototype; `camera.create` adds additional cameras (a security view, a cutscene angle, a render-texture feed).

# capture.cleanup Clears every file from the capture output directory, removing the PNGs left behind by earlier captures. Safe to call when the directory is missing or already empty.

# collage A grid of frames in one image. One picture that answers a question no single frame can: what does this look like from every side, what does this frame look like under each diagnostic pass, what moved over the last three seconds, which of these twenty shots is the shot. ## The axis is always stated A collage varies one thing across its cells, and the argument you pass is what that thing is: | Argument | One cell per | |----------|--------------| | `setups` | whole camera set-up — a shot list, each entry aimed the way a single capture is | | `viewpoints` | named view of one subject — a list, `"front top left"`, or `"sides"` for all six | | `passes` | render pass — `final`, `albedo`, `normal`, `depth`, `motion_vectors`, or any content-registered capture view | | `duration` | sample, spread across that many seconds of gameplay | `setups` and `viewpoints` are one axis — both say where the camera stands — so a request names one of them. Naming a station axis **and** `passes` lays out a 2D grid: a row per station, a column per pass. Reading across a row is one set-up under every pass; reading down a column is one pass from every set-up. A collage with no axis is refused, naming all of them. It does not fall back to time — a silent time default is exactly what taught callers that a collage could only ever be time, so nobody looked for a grid of views. `duration` cannot combine with `setups`, `viewpoints` or `passes`. A cell that differs both in what it looks at and in when it was taken answers neither question; take one collage per axis. ## The shot list Each entry of `setups` is written the way a single capture is aimed — `source`, `camera`, `entity`/`entities`, `position`, `lookAt`, `rotation`, `distance`, `angle`, `margin`, `viewpoint`, `basis`, `projection`, `orthoHeight`, `fov`, `near`, `far` — plus `label` to name the shot on the result. A field an entry leaves out comes from the collage's own options, so a shot list writes down only what makes each shot differ: ```lua zero capture collage --setups '[ { "position": [-14.6, 1.35, 4.5], "lookAt": [-16.3, 1.1, 1.6], "fov": 28, "label": "doorway" }, { "position": [8, 3, 12], "lookAt": [0, 1, 0], "fov": 50, "label": "wide" }, { "entity": "hero", "viewpoint": "front", "basis": "local", "label": "closeup" } ]' ``` Three stations, three lenses, three subjects, one instant, one picture — which is the comparison a framing pass is made of. A key outside that vocabulary is refused where it was written, so a misspelled `fov` names itself instead of quietly framing a shot nobody asked for. ## Reading the result The grid shape is worked out from the number of cells — pass `grid` (`"3x2"`) to state it, and it is an error rather than a silent crop if it cannot hold them. Every cell comes back labelled in `cells`, with its `label`, the `viewpoint` and/or `pass` that made it differ, the `eye` it was rendered from and the lens it was rendered through. A grid is meant to be readable from the response, not guessable from the pictures. Only a time collage reports the window it covered — a station or pass grid samples one instant, and reporting a span for it would invite reading the difference between cells as motion. ## Aiming it Exactly as a single capture is aimed: `entity` (with `viewpoint` and `basis`, or `angle` and `distance`), `position` with `lookAt`, `camera`, or nothing at all for the main scene camera. `projection`, `orthoHeight`, `isolate`, `renderLayers`, `fov`, `width` and `height` all mean what they mean for one frame, and apply to every cell a set-up of its own does not override. The shot most of this exists for is one call: ```lua zero capture collage --entity crate --viewpoints sides --basis local \ --projection orthographic --isolate ``` Six sides of one object, each an orthographic elevation, nothing else in frame.

# feed The newest frame a capture produced on this engine, and who is watching for it. Every capture core publishes here. `latest()` is that frame: a `src` an image widget draws (the render target's guid while the engine still holds it, the written path when the capture produced a file), with the size, the source and the pass beside it. `watch(fn)` calls `fn` as each new frame arrives, and `unwatch(token)` drops the subscription. `shown()` says a viewer is drawing frames from the feed. From then on the feed holds the newest capture's render target so the picture stays up after the capture that made it is finished with it, and lets it go when a later frame replaces it; `shown(false)` stops. With nobody drawing, a capture's target is released the moment its capture is done. The release path asks `retains(rtHandle)` before destroying a target, which is how a held frame survives its own capture. `capture.shared.releaseFrame` is the one place that asks.

# capture.fromCamera Render the scene from a specific named camera's authored pose and settings, returning the picture as an image value (and writing a PNG as well when a `path` is given). By default it uses the camera's exact world-space pose plus its whole lens (fov/near/far, and the projection it builds with the orthoHeight an orthographic camera covers), its diagnostic debugChannel, its post-process setting, its lighting environment (an isolated camera's background and ambient with it) and the scene layers it draws: the camera's own `renderLayers` with the editor's chrome (`EditorUI`) and the authoring overlays (`debug`) taken off, since those are drawn for whoever is editing rather than by that camera into the world. So the frame is precisely what that camera draws of the scene, without changing what is on screen. Pass overrides (`renderLayers`, `debugChannel`, `fov`, `near`, `far`, `projection`, `orthoHeight`, `width`, `height`) to change any of those for the one capture without touching the camera; `renderLayers = "all"` is how the overlays are captured on purpose.

# capture.fromEntity Renders a screenshot orbiting a target entity. The camera position is computed from a distance and a yaw/pitch angle relative to the entity's world position, with an optional offset applied before framing and a configurable field of view and output size. Returns the picture as an image value, and writes a PNG as well when a `path` is given; fails when the entity is missing or its position can't be read. Runs as a background task that waits for the GPU readback, so it works in play mode.

# capture.fromPosition Renders a screenshot from an explicit world-space camera position and returns the picture as an image value. Aim it with a look-at target (defaulting to the origin) or a quaternion rotation that takes precedence, and set field of view, output size, and near/far clip planes. A PNG is written as well when a `path` is given. It spawns an ephemeral camera, triggers one render, reads the frame back, and despawns the camera automatically — running as a background task that waits for the readback, so it works in play mode.

# isolate Show the subject without the other geometry drawn, then put everything back. A prop inside a built scene is behind whatever stands in front of it. The render-layer system already expresses "draw only these" — an entity is a member of layers, a camera renders a filter over them — but it does not remember what it displaced, so every caller wanting one clean shot of one thing hand-rolls the same record, swap and restore. This is that, once, with the restore guaranteed. ## What it changes Exactly one thing: the other geometry is not drawn. Sky, post-process and UI are left exactly as the capture's own render-layer spec stated them, so a final frame stays a final frame. A caller who wants the geometry read flat — no lighting, no atmosphere — asks for a **pass** (`pass = "albedo"`, `pass = "normal"`), which is what passes are for. ## What it does not change Excluded geometry still **casts shadows** onto the subject and still bounces light into it. Lights are not filtered by render layer at all, and the layer mask drives object culling and the geometry passes rather than the shadow passes. This is the render-layer system's deliberate behaviour — a wall stays out of the shot while it still exists for lighting, shadows and physics — so a shadow with no visible caster in an isolated capture is the system working, not a bug. ## The layer One reserved layer name, `captureIsolate`, reused by every isolated capture. Membership is a 32-bit mask with six reserved bits. A layer minted per call would exhaust the namespace in twenty-six captures and leave a trail of empty layers that `renderLayer.list` would show forever. ## Restoring `begin` records each affected entity's membership before moving anything, so the token describes the whole subtree even if the move fails part-way. `restore` puts every entity back to the layers it carried — and an entity that carried **no** explicit membership ends with none, rather than an explicit `default` that looks identical in a picture and different in the data. `restore` is safe to call twice and safe on a token whose subjects have since despawned, which is what lets a caller restore unconditionally on every exit path: success, failure, and once around a whole collage grid rather than between its cells.

# capture.oneshot The low-level one-shot capture primitive. Spawns an ephemeral offscreen camera configured from a source — the viewport, an entity (or several), a world-space position, or a UI window — triggers a single render, and returns the camera and render-target handles synchronously so the caller drives the readback and cleanup itself. This synchronous shape is the only capture path that works from edit mode, where the background-task scheduler the convenience wrappers rely on does not tick. When framing an entity, omitting the distance auto-fits the camera to the union world-space bounds of the target(s), so the subject is guaranteed fully in frame at any scale — a small prop and a large building both fill the shot. Because the data drives the camera, an empty result is a real signal that the geometry or material is wrong rather than a framing mistake; a distance can be passed to force a fixed orbit radius, and an angle sets only the viewing direction. Further options select the debug visualization channel, toggle which layers (scene, sky, post-process, UI) render, and set field of view, clip planes, and output size. The returned camera is left disabled so it can be re-triggered to produce additional frames on the same target, as when stitching a collage.

# shared Module Cross-tool helpers for the `capture` toolbox. Every tool inside this toolbox can pull these in with `require(".shared")`. ## Purpose This module holds the small set of helpers that more than one tool in the toolbox needs — typically the `ok(value, formatted?)` / `fail(message, exitCode?)` constructors that produce the canonical `captureToolResult` shape, plus any toolbox-wide parsers, validators, or shared state. Keep it focused: per-tool implementation stays inside each `<x>.tool/init.luau` (with its own `--!desc` / `--!arg` / `--!return` / `--!example` annotations). ## Usage ```luau local shared = require(".shared") function M.someTool(...) if not ok then return shared.fail("reason", 1) end return shared.ok(value, "") end ``` ## Exports See `init.luau` for the full export surface. Typical exports include `shared.ok(value, formatted?)` and `shared.fail(message, exitCode?)` for building the toolbox's standard result envelope.

# subject What a capture of an entity is a picture of. Every path that frames or isolates an entity (`fromEntity`, `oneshot`, a `setups` or `viewpoints` collage, `isolate`) resolves the name, id or proxy it was given here, so they agree on which entity a name means. ## A player is drawn as its body A player identity (an entity carrying `UserIdentity`) is the registry anchor a connected user is known by. The player is drawn as the body its `avatar` names, a standalone entity at the scene root, so a capture of an identity frames and isolates that body. In play, a player prototype's spawn turns the prototype's copy into the joining user's identity under the prototype's own name and stands the body it names at the spawn point. A capture of the name an author gave their player prototype therefore frames the player's body in play, and the prototype's subtree in edit. ## What a refusal names `describe(target, via)` is the phrase a framing refusal uses for each subject it looked at: the entity's name and id with the number of descendants it holds, and the player identity it is the body of when it was reached through one, or, for a player identity with no live body bound, that the body is missing. | Function | Returns | |----------|---------| | `resolve(t)` | the entity a capture pictures for a name, id or proxy (nil when none matches), and the player identity it was reached through | | `bodyOf(target)` | the live body a player identity is drawn as, or nil | | `isPlayerIdentity(target)` | whether the entity carries `UserIdentity` | | `describe(target, via)` | the refusal phrase for one resolved subject |

# viewpoints The named camera stations a capture can be taken from, and the rule that decides what a name like `front` means. A **viewpoint** is a direction with an image-up: where the camera stands relative to what it is looking at, and which way is up in the resulting picture. Naming one replaces reverse-engineering a `{yaw, pitch}` pair, and names the seven stations that answer "is this built correctly" — the six sides and the corner. | Viewpoint | Sees | |-----------|------| | `front` | the side the subject faces | | `back` | the opposite side | | `right` | the subject's right | | `left` | the subject's left | | `top` | plan view, the subject's front toward the top of the image | | `bottom` | the underside | | `iso` | the corner view where all three axes project equally | An entity faces its local -Z, so the camera that sees its front sits on its +Z. `top` and `bottom` take their image-up from the subject's forward axis, because world up is undefined when you are looking straight down it — that is what puts the subject's front at the top of a plan view instead of leaving the roll to whatever the yaw happened to be. ## Basis `basis` says which axes the name is measured against, and it has **no default**: - **`"local"`** — the subject's own axes. `front` is the side the subject faces, whichever way it is turned, and the frame is fitted to the subject's own extents (`entity:orientedBounds()`). - **`"world"`** — the world axes. `front` is the side facing world +Z, whatever the subject is doing. For a crate turned 40 degrees these are different pictures, so the caller states which one it wants. A default would mean one call site's `front` is the crate's front and another's is the world's, and the difference only shows up in the image. With no subject to take axes from — a capture framed on a world position rather than an entity — both values mean the world axes, and both are accepted. ## Orthographic size `orthoHeightFor` answers the question an orthographic frame asks that a perspective one does not: how big is the frame? A perspective camera backs off until the subject fits; an orthographic camera does not converge, so its size is decided by the subject alone — project the bounds' eight corners onto the camera's up and right axes and take whichever needs more room. Its `half` extents and `axes` must be stated in the same frame. A resolved viewpoint carries both: `frame` axes to use with local-frame extents, and its world axes to use with world extents.

# capture.viewport Renders a screenshot mirroring the active viewport. It mirrors the on-screen camera through the shared capture core, so the frame carries that camera's pose, its lens (`fov`, clip planes and the projection it builds), the lighting environment it states (an `isolated` camera's background and ambient) and the layers it draws the screen with, the editor's own overlays included, so the image matches what's on screen. The raster is the viewport's own pixel size, the size `getViewportSize()` reports. Returns the picture as an image value, and writes a PNG as well when a `path` is given; fails when no active viewport camera exists. Its one argument is a path to also write the PNG to, or a table overriding any of that for one capture: `width` / `height` for a raster of some other size, `renderLayers` for the frame under a different layer spec — `"all !EditorUI !debug"` is the frame without the editor's chrome over it — and `postProcessing` to take it with the grade off. ```lua tools.use("capture", "viewport", "/source/tmp/capture/shot.png") tools.use("capture", "viewport", { path = "/source/tmp/capture/clean.png", renderLayers = "all !EditorUI !debug" }) tools.use("capture", "viewport", { width = 960, height = 540 }) ```

# animation.observe What the engine is posing right now, one line per body. ```lua tools.use("animation", "observe") -- every body in the scene tools.use("animation", "observe", "hero") -- one character ``` Each entry carries: - `animating` — the engine's own verdict, read off the pose it measured landing on the armature rather than off any component field. - `reason` — why a still body is still: `deactivated`, `noRiggedSkeleton`, `noGraph`, `clipUnreadable`, `noOutputNode`, `retargetPending`, `retargetMatchedNoRoles`, `stopped`, `finished`, `paused`, `poseNotApplied`, `poseUnchanged`. - `clip`, `time`, `duration` — the clip driving it and where its playhead is. - `matched` of `total` — how many of the body's bones that clip reaches. A clip from an unrelated rig reads 0 of 50 while its playhead advances. - `bonesWritten`, `poseRevision`, `sinceChangedMs` — the engine's measurement: how much of the pose landed, how many times its values changed, and how long ago they last did. `animation.observe()` in Luau returns the full observation this condenses, and `/zero/runtime/animation` serves the same document as a file.

# animation.play Play one `.animation` clip on an entity's skinned body for a test window, then clean up. ```lua animation.play(animation, entity, opts?) ``` - `animation` — clip identity, name, or `AssetRef<animation>`. - `entity` — entity id, name, or `EntityRef`. The tool finds the first `SkinnedModel` on it or under it (the imported-character layout puts the rigged mesh on a child) and drives that body. - `opts` — optional: - `offset` — seconds (`5`) or a string (`"5s"`, `"500ms"`) to wait before the clip starts. The lookups happen immediately; only the playback is delayed. - `loop` — wrap the clip. Default `false` (play once). - `duration` — seconds to hold the play window. Default: one clip length for a one-shot, or `5` when looping. - `speed` — playback rate multiplier. - `sourceRig` — override the clip's recorded source rig. Returns `{ ok = true, entity, skinnedModel, clip, offset, loop }` once scheduled, or `{ ok = false, error }` when the clip, the entity, or a SkinnedModel under it cannot be found. Playback errors after the offset (an unreadable clip, a rig with no shared roles) are logged, not returned — the lookups are the returned contract. Plays while the world runs (play mode). For durable playback author a `ClipPlayer` component instead; this tool leaves nothing behind and does not disturb an animation the body already runs.

# catalog The controls that come pre-filled for all three devices.

# check Report every problem across a scheme's controls, without activating it.

# control Add one control to a scheme that already exists.

# map Write a whole control scheme — the map and every control it declares.

# inputAuthor shared The catalog every `inputAuthor` tool writes from, and the code that turns a request for a control into the `.inputBinding` folder on disk. ## The catalog A control scheme is mostly the same dozen controls. `move`, `look`, `jump`, `sprint`, `crouch`, `interact`, `attack`, `aim`, `reload`, `menu`, `throttle`, `steer`, `brake` and the rest arrive already covering keyboard and mouse, gamepad and touch, so naming one is enough. Where each lands is a judgement about the device rather than a translation of the keyboard binding: `sprint` is a stick click on a pad because that is where a thumb already rests, and a lower-right button on touch for the same reason. `aim` is a trigger rather than a face button. `from` bases a control on a catalog entry under a different name — a racer's `boost` IS a sprint with another label, and a game's `fire` is an attack — so it keeps the name the game calls it while inheriting three bindings. A name the catalog does not know needs its own `kbm`, `gamepad` and `touch`. The error names which are missing and lists what the catalog would have filled in. ## Refusing a device `false` on a class refuses that device outright, and is the one way to author a control that is deliberately not on a phone or not on a pad. A merely absent class is refused at activation; a refusal is accepted, and carries a reason a player-facing surface can read. ## Shaping an axis An `axis1` / `axis2` takes `deadzone`, `curve`, `invert` and `smoothing`, written into the control so the same numbers apply on every device. A button refuses all four — it is held or it is not, and there is no magnitude to shape. WHEN a button answers is `gate`'s question. ## Gating a control The bindings say WHAT drives a control; `gate` says WHEN it answers at all. A mouse always reports a delta, so a mouse-look with no gate turns the camera whenever the cursor crosses the window — which is why the catalog's `look` carries one and is the only entry that does: ```luau gate = Arming.gate({ rightButton = "free", dragZone = "right", stick = "right" }) ``` `Zin.arming.gate` builds the predicate from the gestures a scheme means by "the player is steering", and the writer emits its require beside the bindings module's. `gate = false` on a request drops an inherited one, for a control that deliberately answers whenever its bindings move. ## What it writes `writeBinding` produces the `<name>.inputBinding/` folder — `init.luau`, `README.md` and `.metadata` — inside the map folder it is given. The generated `init.luau` requires the bindings module through `@builtin::`, which is the spelling that resolves from a world's own source.

# appearance.replaceWithContent Replace one or many entities with a content asset — a bundle or mesh (a generated `.bundle`, a library asset, a primitive). Each target entity and its **whole subtree is removed**, and a fresh spawn of the content takes its slot: same parent, local position, rotation, and name; scale kept unless `keepScale = false`. Components on the removed entities do not carry over — the result is the content, not an upgraded copy of the target. ```lua appearance.replaceWithContent("rider_blockout", "/zero/source/generated/meshes/a_snowboarder.bundle") appearance.replaceWithContent({ "tree_1", "tree_2", "tree_3" }, "@libname::pine.bundle") appearance.replaceWithContent({ "tree_1", "tree_2" }, "@libname::pine.bundle", { dryRun = true }) ``` Targets are entity names or ids — one, a list, or `scene.findAll` records. `dryRun = true` resolves everything and returns the plan (which entities would be removed, how big each subtree is, what would spawn) without mutating, so a batch can be validated before it runs. To swap an entity for another entity that already exists in the scene, use `sc.swap`. To change the mesh on existing entities without touching the entity itself, use `appearance.setModel`.

# appearance.setModel Assign a model (mesh) to the Model and SkinnedModel-bearing entities found by searching an entity hierarchy from a root. The mesh counterpart of setMaterial — it swaps the model field instead of the material. Give the root by id or name and the model as a mesh name, a procedural source like `cube` or `sphere`, or a mesh reference; options let you take only the first match, filter by a name pattern, or exclude the root itself.

# shared Module Cross-tool helpers for the `appearance` toolbox. Every tool inside this toolbox can pull these in with `require(".shared")`. ## Purpose This module holds the small set of helpers that more than one tool in the toolbox needs — the toolbox-wide hierarchy-search and ref-resolution helpers every appearance tool uses. Tools return their raw values (the `tool.assetType` wrapper packages them into the system `ZmToolResult` envelope) and raise via `error(message)` on failure. Keep it focused: per-tool implementation stays inside each `<x>.tool/init.luau` (with its own `--!desc` / `--!arg` / `--!return` / `--!example` annotations). ## Usage ```luau local shared = require(".shared") function M.someTool(material, opts) local rootIds = shared.resolveRoots(opts and opts.roots) local hits = shared.collectAssetRefFields(rootIds, "material", opts) return { roots = rootIds, count = #hits, assigned = hits } end ``` ## Exports See `init.luau` for the full export surface: - `shared.resolveEntity` — one entity id OR name → id. - `shared.resolveRoots` — batch roots argument (one id/name, an array of ids/names/entity proxies, or nil = every scene root) → id array. - `shared.collectAssetRefFields` — schema-driven discovery: walk the targeted hierarchies and return every component public field declared as `Field.assetRef(<category>, ...)` (read from each component's registered schema via `component.getAssetFields`), as `{ id, name, component, field }` hits. - `shared.collectModelEntities` — hierarchy walk collecting the Model / SkinnedModel-bearing entities (backs `setModel`). - `shared.isAssetHandle` — whether a table is shaped like an asset-ref envelope (holds a `__ref` / `guid` / `identity` / `path` string), which is what separates a handle a caller holds from a table of some other kind. - `shared.resolveMaterialRef` / `shared.findMaterial` — material ref resolution; `findMaterial` additionally combs every material in the project for a string argument. - `shared.findShader` — shader ref/identity normalization, combing every shader in the project for a string argument. - `shared.refKey` — stable string key (identity/guid/name/__ref) from a ref value. - `shared.specAcceptsCategory` — `AssetRef<...>` spec-string grammar check, including `AssetRef<a|b>` alternatives.

# appearance.swapMaterials Batch-assign one material across entity hierarchies in a single call — one root, a list of roots, or the whole scene. Targets are discovered from component schemas: every public field declared to hold a material reference receives the material, so Model, SkinnedModel, and every user-authored component carrying a material field participate. Name the material as a word (an exact identity match wins, then an exact leaf name, then the first case-insensitive substring match) or pass a material reference directly. Options let you take only the first assignment or restrict to entities matching a name pattern; each affected entity flashes a blue outline for visual confirmation, and the result lists exactly which fields changed.

# appearance.swapShader Swap the shader on a material while preserving its texture links and property values as far as possible. The change is written to the material asset itself and marks every instance dirty, so it persists and propagates to all entities using that material. Because each shader has its own property and texture vocabulary, values carry across by canonical role — the base-color texture, for instance, maps regardless of what each shader names it — and anything outside the new shader's vocabulary is dropped; the result lists what was carried and what was dropped. Target a single material directly, or target entity hierarchies (one root, a list, or the whole scene): every component field that holds a material reference is read, and every distinct material those fields carry is swapped.

# wld Manage the live world while you work — flip between edit and play, and compose additive scene layers. Flip into play mode — which snapshots the edit state, then starts scripts and physics — and back to edit, which restores that snapshot and stops scripts; query the current mode any time. Load a scene as an additive layer alongside the current one, unload it, or show and hide layers without losing their entities. List the worlds saved on disk, and read the current world state: name, mode, active layer, and available scenes. To list the loaded scene layers or the scenes available to load, use `scene.loadedLayers` and `scene.list`. Saving and loading scene snapshots is the `scene` toolbox (`scene.save` / `scene.load` / `scene.list`).

# guide A guide is a single-file documentation asset: a `<name>.guide/` folder whose body is `guide.md`, surfaced through the `guides` MCP tool, the `man` shell builtin and the Docs panel. Drop one anywhere under `/zero/source/`, a `.package` included, and it shows beside the built-in guides. ## In the Inspector Guide: the guide's word count, its headings, and **Open in Docs**, which opens it in the Docs panel.

# actions A name-registered action registry over `Zin.bindings` + `Zin.context`. A name maps to a list of binding descriptors and one declared context; polling queries (`held`, `pressed`, `released`, `value`, `heldTime`, `repeated`) evaluate those bindings against the live snapshot, and the handler surface (`bind` / `onPressed` / `onReleased` / `onChanged` / `onHeld`) fires the same state from inside `Zin.tick`, after axes and chords advance. A control asset replaced this. A `<name>.inputBinding/` folder inside an `.inputMap/` carries the label, the kind, and a binding for keyboard-and-mouse, gamepad and touch, and a world activates the map and subscribes to the controls it uses: ```luau public = { inputMap = Field.assetRef("inputMap", "@builtin::inputMaps.default", NoSync, "The controls this component reads."), speed = Field.number(5, NoSync, "World units per second at full deflection."), jumps = Field.number(0, NoSync, "How many times the jump control was pressed."), } local controls function awake() controls = public.inputMap:activate() controls.move:onInput(function(v, dt, active) if not active then return end local p = public.entity.position local step = public.speed * dt public.entity.position = { p.x + v.x * step, p.y, p.z - v.y * step } end) controls.jump:onPressed(function() public.jumps += 1 end) end function onDestroy() public.inputMap:deactivate(controls) end ``` `Zin.map.bake(name)` writes the currently active map out as an `.inputMap` asset. `man topics/input` covers the control model. ## Exports Registry: - `M.define(spec: ActionSpec)` — define / replace actions. - `M.remove(name: string)` — remove one action (also tears down handlers). - `M.clear()` — wipe every action + handler. - `M.has(name: string) -> boolean` - `M.names() -> { string }` - `M.get(name: string) -> ActionEntry?` — the registry record behind a name. - `M.active(name: string) -> boolean` — true iff the action's context matches the top of the stack. Polling queries (gated by context): - `M.held(name) / pressed(name) / released(name) -> boolean` - `M.value(name) -> number | Vec2 | 0/1` - `M.heldTime(name) -> number?` - `M.repeated(name, opts?) -> boolean` Event-driven handlers: - `M.bind(name, fn, opts?) -> ActionHandle?` — register a handler. - `M.disconnect(handle: ActionHandle) -> boolean` - `M.onPressed / onReleased / onChanged / onHeld(name, fn, opts?) -> ActionHandle?` - `M.handlerCount(name) -> number` Internal hooks (wired by `zinput/init.luau`): - `M._setAllocator(fn: () -> number)` — shared id allocator. - `M._setEnsureBindingsFn(fn)` / `M._setEnsureLiveFn(fn)` — the cold-read hooks. - `M._owns(handle: number) -> boolean` - `M._dispatchHandlers(firstTickThisFrame: boolean?)` — called by `Zin.tick`. - `M._clearHandlers()` — test isolation. Types: - `ActionEntry = { context: string, bindings: { any } }` - `ActionSpec = { [string]: any }` — define-spec entry shapes. - `ActionState = "Begin" | "Change" | "End" | "Held"` - `BindOpts = { priority: number?, fire: { ActionState }?, once: boolean? }` - `ActionHandle = number` ## Notes - Define-spec shorthand: - `binding` — single binding under the default context. - `{ binding1, binding2 }` — multiple bindings under the default context. - `{ context = "...", binding = ... }` / `{ context, bindings = {...} }` — explicit context. - Context gating is strict: the action's context string must equal `Zin.context.current()`. There is no inheritance up the stack. - Handlers run from `Zin.tick` after axes/chords advance. With no tick host (or `Zin.autoTick.start()`), handlers never fire. - Returning `"sink"` from a handler stops further (lower-priority) handlers on the same action this tick. Anything else continues. - A handler also ends with the context that bound it: one a component binds ends when that component instance goes (removed, hot-reloaded, or its entity despawned), one a module body binds at the module's next run. `scopes.list()` names each as a `luau` row under the context it follows. - `remove(name)` and `clear()` also wipe handlers bound to those actions — handler state never outlives its action. - The handler allocator is swapped at boot (`_setAllocator`) so handles drawn here share one namespace with `Zin.input.on*`.

# arming The predicate a look-style control hands to its `gate`, in the one place it is written. Backs `Zin.arming`. A mouse always has a delta, so an ungated mouse-look turns the camera whenever the cursor crosses the window. The gate answers "is the player steering?" — and every scheme answers it with the same list of gestures in a different combination. `Zin.arming.gate(opts)` takes the combination and returns the predicate. ```luau local Arming = require("@builtin::modules.zinput.arming") return { label = "Look", kind = "axis2", gate = Arming.gate({ dragZone = "right", stick = "right" }), kbm = B.mouseDelta(), gamepad = B.padStick("right", { as = "delta", unitsPerSecond = 1200 }), touch = B.touchDrag({ zone = "right" }), } ``` A locked pointer always arms: a cursor the scheme took is one the player gave to the camera. The rest is named in `opts` — `dragZone` for a finger inside a touch zone, `stick` for a deflected pad stick, `padButtons` for a trigger, `virtualButtons` for an on-screen button. ## The right mouse button `rightButton` is the field the schemes disagree about, because the button means two different things. | Value | Arms while | Used by | |-------|-----------|---------| | `"free"` (default) | the button is down and no other control that can answer on it declares it | the player scheme | | `"held"` | the button is down | the editor scheme, where holding it IS the fly gesture | | `"off"` | never | a control that reads the button through nothing but pointer lock | `"free"` is what lets a world bind the right button to a game action. The engine's default scheme arms mouse-look on that button, and a world that also binds it would otherwise get the camera turn and its own action from one gesture, with no way to decline the camera half short of standing down its whole control group. Under `"free"` the world's binding is the claim: declare the button in a live `.inputMap` control, or in the `Zin.actions` / `Zin.axes` registry, and the arming stands down for as long as that control can answer on it. A control whose map is standing down for a suppressor, or whose context is not the one on top, is one the player cannot reach — the same two terms the tick's own dispatch reads before it evaluates a control — so the button is free of it while that holds and the camera keeps the drag. `Zin.arming.rightButtonClaimants()` names who is holding it, which is the answer to "why did my right-drag stop turning the camera": ```luau Zin.arming.rightButtonClaimants() --> { "map:torch.torchToggle" } ``` The engine's own arming gesture — the `lookDrag` control on the standard scheme, and the `look_drag` action beside it in the registry — is what `"free"` arms FROM, so those two names are passed over by the search. A claim is decided on the frame the button goes down and holds until it comes up. A map activating mid-drag therefore cannot take a gesture away from the camera halfway through it, and the search runs once per press rather than once per gated control per frame.

# autoTick Opt-in per-frame tick loop for zinput. Spawns a `task.spawnSystem` worker that invokes a registered tick function (`Zin.tick`) once per frame, so code-mode snippets and tests without a host script component still drive the stateful subsystems — live binding dispatch, the touch overlay, the emulation floor, axes, chords, rebind, action handlers, event subscribers. `task.spawnSystem` (not `task.spawn`) makes the loop system-owned, so it keeps ticking while the editor pauses gameplay. The first real input read arms this worker on its own. A host script component can also call `Zin.tick(dt)` from its `update(dt)` hook, and `autoTick.start` can be called directly; all three are idempotent alongside each other. ## Exports - `M.start() -> boolean` — spawn the worker if not already running. Returns true when a new loop was started. - `M.stop()` — flag the loop to exit on its next yield. No-op if not running. - `M.isRunning() -> boolean` — current status. - `M._setTickFn(fn: (number?) -> ())` — internal hook. Called once by `zinput/init.luau` to register `Zin.tick`. - `M._setLastTickFrameIdFn(fn: () -> number)` — internal hook. Wires the frame-id accessor so the worker can dedup itself against explicit ticks within the same engine frame. ## Usage ```luau local Zin = require("@builtin::modules.zinput") -- One-shot test / code-mode setup: Zin.autoTick.start() -- ... later ... Zin.autoTick.stop() ``` ## Notes - Single-flight: calling `start()` twice is a no-op — there is at most one worker per VM. - Cross-coroutine dedup: when an explicit `Zin.tick` ran in the same engine frame, the worker skips that frame so `onHeld` handlers and event subscribers don't double-fire. - Module-local `_running` survives Luau require-cache reload, so a re-required `autoTick.module` still reports the loop is live. `restart_instance` resets the VM and clears everything. - `start()` raises if `_setTickFn` hasn't been wired yet — always `require("@builtin::modules.zinput")` first.

# axes A name-registered axis registry over `Zin.bindings`. Any axis or vector binding becomes a `Zin.axes` entry with optional deadzone, exponential smoothing, curve shaping, invert, context gating, and a per-frame gate function. Stateful — `Zin.tick` calls `advance(dt)` each frame to evolve the smoothed reading. A control asset replaced this. An `axis1` / `axis2` `<name>.inputBinding/` inside an `.inputMap/` carries a binding for keyboard-and-mouse, gamepad and touch, so one control reads a stick and a dragging thumb as well as a key, and the binding's own `as` / `unitsPerSecond` / `scale` bring every class into one unit. It declares the same `deadzone`, `smoothing`, `curve` and `invert`, applied by `Zin.scheme` through the same `Zin.utils` maths this reads: ```luau public = { inputMap = Field.assetRef("inputMap", "@builtin::inputMaps.default", NoSync, "The controls this component reads."), speed = Field.number(5, NoSync, "World units per second at full deflection."), jumps = Field.number(0, NoSync, "How many times the jump control was pressed."), } local controls function awake() controls = public.inputMap:activate() controls.move:onInput(function(v, dt, active) if not active then return end local p = public.entity.position local step = public.speed * dt public.entity.position = { p.x + v.x * step, p.y, p.z - v.y * step } end) controls.jump:onPressed(function() public.jumps += 1 end) end function onDestroy() public.inputMap:deactivate(controls) end ``` `Zin.map.bake(name)` writes the currently active map out as an `.inputMap` asset. `man topics/input` covers the control model. ## Exports Lifecycle: - `M.define(spec: AxisSpec)` — define / replace axes. - `M.remove(name: string)` / `M.clear()` — drop one / all axes. - `M.has(name) -> boolean` / `M.names() -> { string }` / `M.get(name) -> AxisDef?` Reads: - `M.value(name: string) -> any` — smoothed value (`number` for scalar, `{ x, y }` for vector). - `M.raw(name: string) -> any` — raw unsmoothed binding value. Tick hook: - `M.advance(dt: number)` — internal; called by `Zin.tick`. - `M._setEnsureBindingsFn(fn)` / `M._setEnsureLiveFn(fn)` — internal cold-read hooks. - `M._resetGateWarnings()` — test helper. Types: - `Curve = string | (number) -> number` — `"linear"` / `"quadratic"` / `"cubic"` / custom. - `AxisDef = { binding, deadzone, smoothing, curve, invert, context, gate, kind }` - `AxisSpec = { [string]: any }` — define-spec. ## Notes - `smoothing` is the exponential time constant in seconds. `0` means "snap to target instantly". Higher = laggier. - `deadzone` is radial for vector bindings — magnitude under the threshold zeros both axes — and scalar for scalar bindings. - Context gating is strict: the axis's declared context must equal `Zin.context.current()`. Mismatch snaps target to 0, and `advance` smooths current down toward 0. - A throwing `gate` falls through to "ungated" for that frame and logs one warning per axis per session (state cleared via `_resetGateWarnings()`). - Without a tick host (or `Zin.autoTick.start()`), `value()` returns the last computed snapshot; `raw()` always reads live.

# back Back, and the claims a world holds on it. Backs `Zin.back`. Back is the platform's own Back: Android's Back button and back gesture, and a keyboard's Back key (`BrowserBack`, which a browser names the same). A world binds any control of its own to it with `Zin.back.press`, which is how Escape or a pad's button becomes Back too. The engine routes every Back press to the first of these that stands: 1. the topmost overlay on screen that states how it is dismissed: an open `popup` carrying `onDismiss`, or an open, dismissible `modal` carrying `onClose`. Its dismissal runs, the way Escape runs it. 2. the claim held that takes it (`Zin.back.claim`: the highest priority, and the most recent among equals): its handler runs. 3. nothing: the platform's own default answers. On Android the app moves to the background; a desktop window and a browser page stay where they are. A claim says ahead of the press that the world takes Back, which is what Android's predictive back asks for: the system decides before a back gesture begins whether it previews going home. Hold a claim while Back means something to the world and release it when Back should leave. Every press is a `back` event in `input.events()`: `{ kind = "back", taker = "overlay" | "claim" | "none" }`. ## Exports - `M.claim(handler, { priority? }) -> Claim`: take Back. Of the claims held, the highest priority (0 when omitted) takes a press no overlay takes, and the most recent among equals. The editor's claim that leaves play is held at -1, under every claim a played world makes. `claim:release()` gives it up, `claim:held()` says whether it is held. A claim a component makes is released when its entity is destroyed. - `M.press()`: press Back, routed as above. - `M.claims() -> number`: how many claims are held. ## Usage ```luau local Zin = require("@builtin::modules.zinput") local claim local function closeMenu() menu.close() if claim then claim:release() claim = nil end end local function openMenu() menu.open() claim = Zin.back.claim(closeMenu) end -- Escape as Back too, through the world's own control. controls.pause:onPressed(Zin.back.press) ```

# bindings Binding primitives and evaluators. A binding names which physical input satisfies a control on one device class — the value a `.inputBinding` record's `kbm` / `gamepad` / `touch` field holds. A binding is a plain descriptor table (e.g. `{ kind = "key", code = "KeyW" }`); the evaluators (`evalHeld`, `evalPressed`, `evalReleased`, `evalAxis`, `evalVector`) dispatch on `kind` and read `Zin.state` / `Zin.virtual` to produce the live signal. `Zin.scheme` runs them once per tick for every live binding, and they are public so chords, axes and library authors can evaluate a binding without reimplementing the dispatch. ```luau -- boost.inputBinding/init.luau local B = require("@builtin::modules.zinput.bindings") return { label = "Boost", kind = "button", kbm = { B.key("ShiftLeft") }, gamepad = { B.padButton("left_shoulder") }, touch = { B.touchButton({ zone = "right-lower" }) }, } ``` ## Two frames, and which binding speaks which - **Stick bindings** (`vector`, `touchStick`, `padStick`) report the direction the player *means* — forward is `+y`. A `move` control carrying a keyboard class and a stick class drives both devices the same way. - **Delta bindings** (`mouseDelta`, `touchDrag`, `scroll`) report movement that happened — down is `+y`, because that is what the pointer did. They also differ in magnitude: a stick is a position bounded at 1.0, a delta is however far the device moved this frame. A control carrying both — look, zoom — declares `as = "delta"` on its held class with the `unitsPerSecond` a full deflection is worth, and `scale` on a delta class that counts in its own units (a pinch in pixels beside a wheel in notches). `Zin.scheme` applies both, so one sensitivity covers every device. ## Exports Keyboard and mouse constructors: - `M.key(code: string) -> Binding` - `M.mouse(button: string) -> Binding` — `"left"` / `"right"` / `"middle"`. - `M.modKey(spec: string) -> Binding` — `"Ctrl+S"`, `"Shift+Alt+P"`, `"Escape"`. - `M.axis(plus: Binding, minus: Binding, opts: StickOpts?) -> Binding` - `M.vector(right, left, up, down: Binding) -> Binding` - `M.wasd() -> Binding` - `M.arrowKeys() -> Binding` - `M.mouseDelta(axis: string?) -> Binding` — this frame's mouse delta. - `M.scroll(axis: string?) -> Binding` — scrollwheel delta. - `M.pointerLocked() -> Binding` — boolean, true while locked. Touch constructors (fed by `Zin.virtual`, which the on-screen controls write): - `M.touchStick(opts: { zone, axis? }) -> Binding` — vector. - `M.touchButton(opts: { zone?, label?, icon?, priority?, size?, group? }) -> Binding` — boolean. - `M.touchDrag(opts: { zone, axis? }) -> Binding` — per-frame delta vector. - `M.touchPinch(opts: StickOpts?) -> Binding` — scalar pinch delta. Gamepad constructors (addressed by canonical position; `slot` names one pad, omitting it means any connected pad): - `M.padButton(button: string, slot: number?) -> Binding` - `M.padAxis(axis: string, opts: StickOpts?) -> Binding` - `M.padStick(stick: string, opts: StickOpts?) -> Binding` - `M.padTrigger(trigger: string, slot: number?) -> Binding` Predicates: - `M.isBoolean(b) -> boolean` - `M.isAxis(b) -> boolean` - `M.isVector(b) -> boolean` Evaluators: - `M.evalHeld(b) -> boolean` - `M.evalPressed(b) -> boolean` - `M.evalReleased(b) -> boolean` - `M.evalAxis(b) -> number` - `M.evalVector(b) -> Vec2` Pretty-printers: - `M.format(b: any) -> string` — single binding ("Space", "Ctrl+S", "WASD"). - `M.formatAll(bindings: any?, sep: string?) -> string` — joined summary; non-table arg returns `"<no bindings>"`. Types: - `Binding = { [string]: any }` — descriptor table with `kind` plus type-specific fields. - `Vec2 = { x: number, y: number }` - `StickOpts = { slot: number?, as: string?, unitsPerSecond: number?, scale: number? }` ## Notes - Bindings are pure data tables — they serialize cleanly, compare with `Json.encode`, and have no metatables. - `modKey` normalises modifier order in `format()` so a key written as `Shift+Ctrl+KeyS` displays the same as `Ctrl+Shift+KeyS`. The internal `mods = { ctrl, shift, alt }` flags are the canonical representation. - Mouse-delta, scroll, drag and pinch values are raw physical deltas, not clamped to `[-1, 1]`. `as = "delta"` + `unitsPerSecond` on the stick class beside them is what puts every class in one unit. - `as = "delta"` needs `unitsPerSecond` and has no default: the number that makes a pad feel right is the point of declaring the conversion, and a default would be right for one control and quietly wrong for the next. - Reading a member this module does not have raises immediately, naming the nearest members and the full list — a wrong guess at a constructor surfaces at the call rather than as `attempt to call a nil value` later. - `format()` always returns a string, never throws — returns `"<invalid>"` for malformed input.

# chords Chord and sequence state machines. Two kinds: - `sequence` — ordered steps with a per-step window and an overall window (Konami-style codes). - `simultaneous` — N bindings pressed inside a tolerance window, order-insensitive (piano-chord / fighter-combo style). Chords are edge-triggered — `fired(name)` is `true` for exactly one frame when the full pattern completes. A continuous combo like "Ctrl+S held" is a `button` control whose `kbm` class is `B.modKey("Ctrl+S")`: a chord recognizes a pattern over time, which is a different question from whether a combination is down right now. ## Exports Lifecycle: - `M.define(spec: ChordSpec)` — define / replace chords. - `M.remove(name) / M.clear()` - `M.has(name) -> boolean` / `M.names() -> { string }` / `M.get(name) -> ChordDef?` Queries: - `M.fired(name) -> boolean` - `M.progress(name) -> number` (0..1) - `M.cancel(name)` — reset state. Subscriptions: - `M.onFire(name, callback, opts: OnFireOpts?) -> OnFireHandle` - `M.off(handle: OnFireHandle)` Tick hooks (called by `Zin.tick`): - `M._beginTick()` — clears per-frame fire flags. - `M._observeEvent(ev)` — advance state machines on an input event. - `M._advanceTime(dt)` — age out stale recognizers. - `M._dispatchFires()` — invoke onFire callbacks for chords that fired this tick. Types: - `ChordDef = { kind, context, steps?, stepWindowSec?, windowSec?, bindings?, toleranceSec? }` - `ChordSpec = { [string]: any }` - `OnFireOpts = { context: string?, once: boolean? }` - `OnFireHandle = { id: number, name: string, off: () -> () }` ## Usage ```luau local Zin = require("@builtin::modules.zinput") local B = Zin.bindings Zin.chords.define({ konami = { kind = "sequence", steps = { B.key("ArrowUp"), B.key("ArrowUp"), B.key("ArrowDown"), B.key("ArrowDown"), B.key("ArrowLeft"), B.key("ArrowRight"), B.key("ArrowLeft"), B.key("ArrowRight"), B.key("KeyB"), B.key("KeyA") }, stepWindowSec = 0.5, windowSec = 4.0, }, }) Zin.chords.onFire("konami", function() unlockBonus() end) ``` ## Notes - Without a tick host (or `Zin.autoTick.start()`) chords never fire — every state machine is dormant until `Zin.tick` drives it. - A wrong key in a sequence resets to step 0. If the wrong key happens to match step 1, the recogniser restarts from step 1 on the same event (so a long input run can self-recover). - For simultaneous chords, the tolerance window starts on the first matched press. If all bindings haven't matched before `toleranceSec` elapses the state resets — `firstPressAt` is reset too so the next press starts a fresh window. - Subscriber errors are caught and logged via `log.warn`.

# zinput / clock The wall-clock time source for zinput's time-window features — held-time and synthetic key repeat (`state`), chord step / total windows (`chords`), and the tick's fallback delta (`init`). `nowSeconds()` reads the engine's per-frame `getTime()` (seconds since engine start, published from the native clock), which advances with wall time on every platform. `os.clock` measures CPU time and barely moves across scheduler yields on wasm, so it is only the fallback for contexts with no engine time surface. ## API - `nowSeconds() -> number` — current wall-clock seconds for input timing. - `_setClock(fn: (() -> number)?)` — internal; override the clock with `fn` (returns seconds) or `nil` to restore the default. For deterministic timing: fixed-step replay and tests that need a controllable clock.

# conflicts Binding-conflict detection — reverse index over the `Zin.actions` / `Zin.axes` registries. Two bindings "conflict" when they map to the same canonical leaf key (e.g. both bind `Space`, or an action binds `Space` and an axis's `plus` side is `Space`). Backs `Zin.conflicts`. Rows carry a `kind = "real" | "intentional"` classification: two or more action owners in the *same context* are `real`; action + axis overlaps, cross-context actions, and explicitly marked pairs are `intentional`. `realOnly()` is the noise-filtered view the Input tab uses; `forActiveProfile()` is the full audit list. ## Exports Marking: - `M.markIntentional(actionA: string, actionB: string) -> (boolean, string?)` - `M.unmarkIntentional(actionA: string, actionB: string) -> boolean` - `M.isIntentional(actionA: string, actionB: string) -> boolean` - `M.listIntentional() -> { { string } }` Probing: - `M.bindingKey(b: any) -> string?` — `"key:Space"`, `"modKey:Ctrl+KeyS"`, `"mouse:left"`, or `nil`. - `M.find(binding: any) -> { Owner }` — owners that share any leaf key with the probe. - `M.forActiveProfile() -> { ConflictRow }` — every shared-key row. - `M.realOnly() -> { ConflictRow }` — subset where `kind == "real"`. Internal (wired by `zinput/init.luau`): - `M._setPersistFn(fn)` — registers the save-back closure. - `M._loadIntentionalPairs(pairs_)` — seed on profile activate. - `M._serializeIntentional() -> { { string } }` — for profile save. - `M._resetIntentional()` — test isolation. Types: - `Owner = { target: string, name: string, slot: any }` — `target` is `"action"` or `"axis"`. - `ConflictRow = { key: string, owners: { Owner }, kind: string }` — `kind` is `"real"` or `"intentional"`. ## Usage ```luau local Zin = require("@builtin::modules.zinput") -- Audit view (e.g. inside the Input tab): for _, row in ipairs(Zin.conflicts.forActiveProfile()) do print(row.key, row.kind, #row.owners) end -- Suppress an intentional same-context overlap: Zin.conflicts.markIntentional("crouch", "slide") ``` ## Notes - `modKey` leaf keys list modifiers alphabetically (`Alt+Ctrl+Shift`), so `Shift+Ctrl+KeyS` and `Ctrl+Shift+KeyS` compare equal. - Continuous-source bindings (`mouseDelta`, `scroll`, `pointerLocked`) never appear in conflict reports — two axes reading `mouseDelta` is by-design. - Persistence: `_setPersistFn` is wired by `zinput/init.luau` so marks flow into the active profile JSON without conflicts needing a hard dep on profile. - The classifier counts every same-context action pair. If even one pair is unmarked, the row is `"real"`; otherwise `"intentional"`.

# context The input context stack. Pushing a named context ("ui", "menu", "vehicle", ...) stands down everything whose declared context doesn't match the top of the stack; popping restores the previous mode. The bottom is permanently `"default"` so unbalanced pops never leave the stack empty. A control, an action, an axis or an event subscriber can declare one context. `Zin.scheme` honours it on every live binding, so a vehicle-context Brake neither fires nor draws its touch button while the default context is on top. A map standing another map's `group` down is the coarser instrument beside this one: `group` / `suppresses` on an `.inputMap` decides which whole schemes answer, while a context decides which entries within a live scheme do. ## Exports - `M.default() -> string` — the always-on context name (`"default"`). - `M.push(name: string)` — push a context. No-op when `name` is not a non-empty string. - `M.pop() -> string?` — pop the top context; returns `nil` when only `default` remains. - `M.current() -> string` — name at the top of the stack. - `M.contains(name: string) -> boolean` — true if `name` is anywhere in the stack. - `M.stack() -> { string }` — a copy of the live stack, bottom→top. - `M.reset()` — pop everything except `default`. - `M.with<T...>(name: string, fn: () -> T...) -> T...` — run `fn` with `name` pushed, pop guaranteed. ## Usage ```luau local Zin = require("@builtin::modules.zinput") -- menuClose.inputBinding/init.luau declares `context = "ui"`, so it reads -- only while "ui" is on top. Zin.context.push("ui") Zin.context.current() -- "ui" Zin.context.pop() Zin.context.with("ui", function() -- inside this block the "ui" controls are the ones that read end) ``` ## Notes - The stack lives in the module's upvalues, so it persists across Luau script runs that share the require cache. A `restart_instance` resets it (matches the standard caching contract). - Pushing the same context twice requires two pops to fully remove — matches typical UI nesting (a sub-menu pushes "ui" again). - A declared context is compared strictly equal to `current()`; there's no inheritance up the stack.

# controllers Reusable controller factories on top of the `Zin.actions` handler, `Zin.context` and `Zin.state` surfaces. Each factory returns a stateful controller object with the `:start / :stop / :update / :isActive` lifecycle and a small set of intent readers. `free`, `orbit` and `fps` read keyboard and mouse, and their `gamepadIndex` opt is accepted and ignored. `vehicle` reads an input map, so every control answers to keyboard, gamepad and touch. ## Exports Factories (module-level): - `M.free(opts: FreeOpts) -> FreeController` — free-fly designer/debug camera (WASD plane + Q/E vertical + Shift sprint + mouse-look). Writes the entity transform directly. - `M.orbit(opts: OrbitOpts) -> OrbitController` — orbit camera around a target (drag-rotate, scroll-zoom, optional pan). - `M.fps(opts: FpsOpts) -> FpsController` — first-person walker. Produces motion intent for the host (`getDesiredVelocity`, `consumeJumpRequest`, `isSprinting`, `isCrouching`); convenience `applyToTransform` writes directly. - `M.vehicle(opts: VehicleOpts?) -> VehicleController`: driving intent for a ground vehicle, read from `opts.inputMap` (an inputMap ref or its path; `@builtin::inputMaps.vehicle` by default): `getThrottle` (-1..1, positive forward), `getSteer` (-1..1, positive right), `getBrake` (0..1), `isHandbrakeHeld`, `isBoosting`, `consumeRecoverRequest` (true once per press), `consumeZoom` (wheel notches since the last read). `:start` activates the map, which stands the walking scheme down for as long as it is live; `:stop` releases it. The `VehicleController` component drives a car's `WheelCollider`s from this intent. Controller-object methods (returned by the factories; called with `:`): - `:start()` — define actions, push context, optionally request pointer-lock. - `:stop()` — drop actions, pop context, release pointer-lock if `:start` took it. - `:isActive() -> boolean` - `:update(dt: number)` — per-frame; no-op when context doesn't match. - `:actionNames() -> { string }` - `FreeController`: `:setMoveSpeed`, `:setLookSensitivity`, `:getCameraYaw`, `:getCameraPitch`, `:getVelocity`. - `OrbitController`: `:getDistance`, `:getYaw`, `:getPitch`, `:setDistance`, `:setTarget`. - `FpsController`: `:setMoveSpeed`, `:setLookSensitivity`, `:getCameraYaw`, `:getCameraPitch`, `:getDesiredVelocity`, `:isSprinting`, `:isCrouching`, `:consumeJumpRequest`, `:applyToTransform`. Types: - `FreeOpts`, `OrbitOpts`, `FpsOpts` — opts tables; `entity` is the only required field. - `VehicleOpts`: `{ inputMap?, label? }`; every field is optional. - `FreeController`, `OrbitController`, `FpsController` — opaque (typed as `any` so the metatable plumbing doesn't fight the LSP). ## Usage ```luau local Zin = require("@builtin::modules.zinput") local cam = Zin.controllers.free({ entity = cameraEntityId }) cam:start() -- per frame: Zin.tick(dt) cam:update(dt) ``` ## Notes - Each controller's actions are namespaced under `actionPrefix` (default `"free_fly."`, `"orbit."`, `"fps."`). Two co-existing controllers with overlapping prefixes will trample each other — override the prefix when stacking. - The context push/pop pair makes `Zin.context.push("ui")` cleanly suppress every controller without `:stop`. The controller's `:update` early-outs when `Zin.context.current()` doesn't match. - Rotation convention: yaw=0, pitch=0 → `(0, 0, -1)` (-Z forward). Mouse Δx adds to yaw; mouse Δy subtracts from pitch (invert with `invertY = true`). Pitch is clamped to `[-π/2 + ε, π/2 - ε]`. - `FpsController` produces *intent* — gravity, collision, and the jump impulse are the host's responsibility. The `consumeJumpRequest()` accessor returns `true` once per press, so the physics step can fire its impulse exactly once. - Free / FPS request pointer-lock at `:start` (opt out with `requestPointerLock = false`) and only release it on `:stop` when they were the ones who acquired it.

# emulation Map-driven input-emulation floor. Reads `Zin.virtual` (the store on-screen touch controls write) each tick and emulates the equivalent kbm keys / mouse delta, so raw-key-polling worlds (`Zin.state.keyDown`) move from touch input without themselves reading `Zin.virtual` or `Zin.bindings`. Backs `Zin.emulation`. ## What it derives, per tick, from the active effective map Only entries whose `context` matches the current input context (`Zin.context.current()`) participate — the same activity rule `Zin.actions` applies, so a vehicle-context Brake button emulates nothing while the default context is on top. - **touchButton actions** — the OR of every touchButton binding in an action's touch class drives its first kbm `key` binding's code, emulated on the held state's transitions. Actions with no kbm key binding are skipped. - **touchStick axes** whose kbm class is a `vector` of four `key` arms (the `B.wasd()` / `B.arrowKeys()` shape) — each stick component thresholded at `+-0.5` drives the matching direction key. The stick's y is screen-down positive (`Virtual.setStick`'s convention), so negative y (stick pushed up) drives the vector's `up` arm and positive y its `down` arm. - **touchDrag axes** whose kbm class is `mouseDelta` — the drag delta, scaled by `M.setSensitivity`, emulated via `input.emulateMouseDelta` whenever it's non-zero. A control that is already a live binding (`Zin.scheme.has(name)`) is skipped: its touch class drives it directly, and synthesising a key for it as well drives it twice, from opposite conventions. The judgement is per control, because a world can subscribe to some things and poll raw keys for others. Emitted keys are tracked locally and diffed each tick — a held key is emulated once, on the press transition, never re-emitted every tick a control stays held. Switching the active map (name change between ticks) or the input context (top-of-stack change) releases everything emitted under the previous map/context. ## Exports - `M.setSensitivity(n: number)` / `M.getSensitivity() -> number` — the touchDrag -> mouse-delta multiplier (default `1.0`). - `M._advance()` — internal: one tick of the floor, called by `Zin.tick`. - `M._reset()` — test-only: release everything emitted, clear state. ## Usage ```luau local Zin = require("@builtin::modules.zinput") Zin.map.ensureActive() Zin.emulation.setSensitivity(1.5) -- Zin.tick(dt) advances the floor; a virtual Jump press then reads as -- Zin.state.keyDown("Space") for any world that only polls kbm state. ```

# events Frame-event readers and edge-event subscriptions. Wraps the internal `__zero_input.events()` FFI with per-kind filters and a push-style `on/off` API that fires from `Zin.tick`. Backs `Zin.events`. `__zero_input.events()` returns the current frame's events in dispatch order; reading does not consume them. Subscription dispatch happens during `Zin.tick` (or `Zin.autoTick.start()`'s worker), so without a tick host `on()` callbacks never fire. ## Exports - `M.frame() -> { InputEvent }` — this frame's events in dispatch order. - `M.iter() -> iterator` — `ipairs` over `frame()`; reads better at call sites. - `M.keysDown / keysUp / mouseDowns / mouseUps / mouseMoves / mouseWheels / texts() -> { InputEvent }` — per-kind filters. - `M.on(filter: EventFilter, callback, opts: SubOpts?) -> SubHandle` — register an edge subscriber, returns handle. - `M.off(handle: SubHandle | number)` — cancel a subscriber. - `M.clearSubscribers()` — drop all subscribers (test isolation). - `M.subscriberCount() -> number` — current subscriber count. - `M._dispatch(events)` — internal; called by `Zin.tick`. Types: - `InputEvent = { [string]: any }` — one entry of the frame event log. - `EventFilter = string | { [string]: any } | (any) -> boolean` - `SubOpts = { context: string?, once: boolean?, priority: number? }` - `SubHandle = { id: number, off: () -> () }` ## Usage ```luau local Zin = require("@builtin::modules.zinput") for _, ev in ipairs(Zin.events.frame()) do if ev.kind == "key.down" and ev.code == "Space" then -- jump on the press edge, not on every frame the key is held end end local h = Zin.events.on({ kind = "key.down", code = "Escape" }, function(ev, gpe) if not gpe then closeMenu() end end, { context = "ui" }) -- later: h.off() ``` ## Notes - Reading `frame()` is non-destructive — multiple consumers can iterate the same frame's events. The log clears at end-of-frame and is rebuilt next frame. - Subscribers dispatch in priority-desc order; equal priorities preserve insertion order. Returning `"sink"` from a callback halts further dispatch for that single event (lower-priority subscribers skipped). - `once = true` subscribers auto-unsubscribe after their first matching dispatch. - A subscription ends with the context that made it as well as with `off`: one a component makes ends when that component instance goes (removed, hot-reloaded, or its entity despawned), one a module body makes at the module's next run. `scopes.list()` names each as a `luau` row under the context it follows. - Subscriber errors are caught and logged via `log.warn`; one bad handler doesn't break dispatch.

# gamepad The pad-facing surface: which controllers are connected, what family they belong to, and what a legend should call a canonical button on them. Backs `Zin.gamepad`. Bindings never come through here. A scheme binds `B.padButton("south")` and evaluates against whichever pad is connected; this module answers the questions a UI asks — how many players have controllers, and which letter to draw on a prompt. ## Position, not letter The canonical layout names a button by POSITION, because position is what every pad shares — `south` is the lower face button on all of them. The letter printed on it is not shared: `south` is `A` on an Xbox pad, `Cross` on a PlayStation one, and `B` on a Nintendo one. `label()` resolves that from the device name the platform reported, so a binding stays device-independent and only the drawn glyph changes. Anything unrecognised reads as `xbox`, because the standard mapping every backend normalises to is the Xbox layout. ## Exports - `M.list() -> { Pad }` — every connected pad, in slot order. - `M.count() -> number` — how many pads are connected. - `M.get(slot) -> Pad?` — the pad in one slot. - `M.family(slot?) -> string` — `"xbox"` | `"playstation"` | `"nintendo"`. - `M.label(button, slot?) -> string` — the legend for a canonical button. - `M.available() -> boolean` — whether the session has ever seen a pad. Latched by the first connection, so unplugging one does not flip a scheme's prompts back to keyboard glyphs on a cable knock. Types: - `Pad = { slot, name, buttons, buttons_pressed, buttons_released, axes, simulated }` — the button fields are arrays of canonical names, `axes` a map of canonical axis name to number, and `simulated` marks a pad that came from the `inputSim` toolbox rather than a device. ## Usage ```luau local Zin = require("@builtin::modules.zinput") if Zin.gamepad.count() >= 2 then startLocalCoop() end ui.text("Press " .. Zin.gamepad.label("south") .. " to jump") ```

# gestures Touch-gesture recognizers over the touch event stream + contact snapshot; backs `Zin.gestures`. Discrete gestures (tap, double-tap, long-press, swipe) fire subscriptions; continuous gestures (pinch, pan) expose per-tick deltas computed from the two-finger contact state. Advanced by the `Zin.tick` pipeline — dormant until ticked. ## Discrete vs continuous - **Discrete** (tap / double-tap / long-press / swipe) — recognized on a release or hold-duration edge, delivered via `on*` subscriptions on the tick they fire. - **Continuous** (pinch / pan) — recomputed every tick from the two-finger contact positions; read via `pinchDelta()` / `panDelta()`, reset to zero at the start of each tick. The pair is formed from the contacts no on-screen widget is working, the ones `Zin.touchControls` says its stick and buttons hold having been left out, so a thumb on a movement stick is no part of a pinch. ## Config `Zin.gestures.configure` overrides recognition thresholds (unspecified fields keep their current values): | Field | Default | Meaning | |---|---|---| | `tapMaxDuration` | `0.25` s | press→release within this = tap candidate | | `tapMaxMovement` | `16` px | total movement below this = tap candidate | | `doubleTapWindow` | `0.3` s | two taps within this = double-tap | | `longPressDuration` | `0.5` s | held past this (still) = long-press | | `swipeMinDistance` | `60` px | release travel at/above this = swipe candidate | | `swipeMinVelocity` | `800` px/s | release velocity at/above this = swipe | ## Exports - `M.configure(opts: { [string]: number })` — override recognition thresholds. - `M.getConfig() -> GestureConfig` — current thresholds (a copy). - `M.onTap(fn) -> number` — subscribe to taps; `fn(ev)` with `ev = { x, y, id, duration }`. - `M.onDoubleTap(fn) -> number` — subscribe to double-taps; `ev = { x, y, id }`. - `M.onLongPress(fn) -> number` — subscribe to long-presses (fires once per contact); `ev = { x, y, id }`. - `M.onSwipe(fn) -> number` — subscribe to swipes (fires on release); `ev = { direction, dx, dy, velocity, id }`. - `M.off(handle: number) -> boolean` — unsubscribe a handle from any `on*` function. - `M.pinchDelta() -> number` — this tick's two-finger pinch delta in px (positive = spreading). - `M.panDelta() -> (number, number)` — this tick's two-finger pan delta `(dx, dy)` in px. - `M._beginTick()` — internal: clear per-tick continuous deltas; wired by `Zin.tick`. - `M._observeEvent(ev: any)` — internal: builds contact tracks and classifies discrete gestures on release; wired by `Zin.tick`. - `M._advanceTime(dt: number)` — internal: time-based recognition (long-press) + continuous deltas; wired by `Zin.tick`. - `M._setWidgetContacts(ids: { [number]: boolean })` — internal: the contacts an on-screen widget is working; published each tick by `Zin.touchControls`. - `M._pairDown() -> boolean` — internal: whether two contacts no widget is working are down, which is the pair a two-finger gesture reads. - `M._dispatchFires()` — internal: deliver this tick's discrete gesture fires; wired by `Zin.tick`. - `M._reset()` — test-only: clear all tracks, fires, deltas, and subscriptions. Types: - `GestureConfig = { tapMaxDuration, tapMaxMovement, doubleTapWindow, longPressDuration, swipeMinDistance, swipeMinVelocity }` (all `number`) ## Usage ```luau local Gestures = require("@builtin::modules.zinput.gestures") Gestures.onTap(function(ev) print("tap at", ev.x, ev.y) end) Gestures.onSwipe(function(ev) if ev.direction == "left" then nextPage() end end) -- inside a per-frame update: local zoom = Gestures.pinchDelta() local dx, dy = Gestures.panDelta() ``` ## Notes - Pinch and pan deltas are zero unless exactly two contacts are down this tick that no on-screen widget is working. - Long-press fires once per contact — the contact must stay within `tapMaxMovement` for the full `longPressDuration` and is marked so it never also fires a tap on release. - Every qualifying release fires `tap`; the second tap of a double-tap fires `doubleTap` and `tap` on the same tick. - Handler errors during dispatch are caught and logged, not propagated.

# input Event-driven listeners for whole categories of raw input, on top of `Zin.events.on`. Sugar for the most-common filter shapes (`onBegan` / `onEnded` / `onChanged` / `onTextInput`) plus a `lastInputType()` accessor. Backs `Zin.input`. These sit beneath the control model: they answer "any key went down" rather than "the Jump control fired", which is what rebind capture, dev tooling and text-adjacent UI need. Gameplay subscribes to controls through an activated map — `man topics/input` covers that path. ## Exports - `M.lastInputType() -> string?` — userInputType of the most recent input event (`"Keyboard"` / `"Mouse"` / `"Touch"` / `nil`). - `M.onBegan(fn, opts?) -> Handle` — fires on `key.down` / `mouse.down`. - `M.onEnded(fn, opts?) -> Handle` — fires on `key.up` / `mouse.up`. - `M.onChanged(fn, opts?) -> Handle` — fires on `mouse.move` / `mouse.wheel`. - `M.onTextInput(fn, opts?) -> Handle` — fires on committed text input; handler signature is `(text, gpe)`. - `M.disconnect(handle: Handle) -> boolean` — tear down a subscription; idempotent. - `M.subscriberCount() -> number` — number of active composite subscriptions. - `M._reset()` — test-only: clear `lastInputType`. - `M._disconnectAll()` — test-only: tear down every subscription this session. - `M._observeEvent(ev: any)` — internal observer driven by `Zin.tick`. - `M._setAllocator(fn: () -> number)` — internal: wire the shared handle-id allocator. - `M._owns(handle: number) -> boolean` — internal: handle-ownership check for cross-API disconnect. Types: - `Filter = string | { [string]: any } | (any) -> boolean` - `SubOpts = { priority: number?, context: string?, once: boolean? }` - `Handle = number` — composite-handle id, drawn from a shared allocator with `Zin.actions`. ## Usage ```luau local Input = require("@builtin::modules.zinput.input") local h = Input.onBegan(function(io, gpe) if io.kind == "key.down" and io.code == "Space" then print("jumped") end end, { priority = 10 }) Input.disconnect(h) ``` ## Notes - Handler signatures: `fn(io, gpe)` for `onBegan` / `onEnded` / `onChanged`; `fn(text, gpe)` for `onTextInput`. `gpe` is the event's own `consumed_by_ui` marker, stamped as the event arrived against the focus the UI holds over that event's surface: pointer focus for a mouse or touch event, keyboard focus for a key or text event, and never for a pad. - Return `"sink"` from a handler to stop further (lower-priority) handlers receiving the same event. - A single `on*` call installs N underlying `Zin.events.on` subscribers but returns ONE composite handle — `disconnect` tears them all down together. - A composite handle also ends with the context that made it: one a component registers ends when that component instance goes (removed, hot-reloaded, or its entity despawned), one a module body registers at the module's next run. `scopes.list()` names each as a `luau` row under the context it follows. - Handle ids share a namespace with `Zin.actions` (`_setAllocator` wires this at boot), so `Zin.actions.disconnect` accepts handles returned here. - `lastInputType` is updated by the tick observer; text events do not bump it (the preceding key.* event already did).

# map A record-shaped map: `Zin.actions` / `Zin.axes` entries organized per device class (`kbm` | `gamepad` | `touch`) with single-parent `extends`, materialized into an EFFECTIVE map and applied as one binding set at a time. Backs `Zin.map`. An `.inputMap` asset with `.inputBinding` children is the map, and `Zin.scheme` is the live set: several maps are live at once and compose, each declaring the `group` its controls belong to and the groups it `suppresses` while live. `Zin.map.bake(name)` writes the map active here out as an `.inputMap` asset, with synthesis made explicit and editable. `man topics/input` covers the control model. Resolution order per action/axis and device class: 1. the map's own class bindings 2. the `extends` parent's (recursively) 3. synthesis from the `kbm` shape (touch class only) A vector `kbm` binding synthesizes a `touchStick`; a boolean `kbm` binding synthesizes a `touchButton`; a vector-shaped `mouseDelta` synthesizes a `touchDrag`; `scroll` synthesizes a `touchPinch`; an `axis`-kind `kbm` binding and a single-axis `mouseDelta` are left unsynthesized (ambiguous which touch shape to guess). A synthesized `touchButton` carries `priority = 100` — an authored `B.touchButton`'s default priority (50) always outranks synthesis on the touch-controls overlay, so a hand-tuned scheme's own layout hints win over whatever filled a gap automatically. A profile record (actions with `bindings`, axes with `binding`) is accepted anywhere a map record is — its bindings are treated as the `kbm` class. ## Activation and the `<axis>@<class>` convention `activate` flattens every class's bindings into each action/axis's single bindings list and applies it through `Zin.profile`, so any device fires the action. For axes — which evaluate exactly one binding — the primary slot is the first of `kbm` / `touch` / `gamepad` present; any OTHER class's binding registers as a class-suffixed sibling axis (`look` + `look@touch`). A consumer that wants both device values reads both names; a kbm-only world's primary axis (`Zin.axes.value("move")`) behaves identically whether or not a touch class exists. `Zin.profile.ensureActive("default")` delegates to `Zin.map.ensureActive()` when the resolved scheme is the builtin default, so it carries every device class rather than kbm alone. `Zin.map.ensureActive()` / `Zin.map.activate` win regardless of call order (`activate` re-registers the flattened profile under the same name and forces a reactivate). ## Exports - `M.materialize(record: any) -> effective map` — resolve `extends` + synthesize, without activating. - `M.activate(record: any) -> effective map` — materialize + apply as the live binding set (registers/activates a `Zin.profile`). - `M.effective() -> effective map?` — the active map's materialized form. - `M.activeName() -> string?` — the active map's name. - `M.bindingsFor(eff, name, class) -> bindings?` — one action/axis's bindings for one device class. - `M.ensureActive() -> effective map` — keep the current map, else activate `@builtin::inputMaps.default`. - `M.bake(name: string) -> AssetRef` — write the active effective map as a new `.inputMap` instance (`asset.create("inputMap", name, { record = ... })`). - `M.addTouchButton(actionName, buttonOpts, emitKey?) -> binding` — add a touchButton binding to an action's touch class on the live effective map (creating the action if absent) and re-activate. Used by `Zin.touchControls.button`. - `M.removeTouchButton(actionName, binding) -> boolean` — remove a binding added via `addTouchButton` and re-activate; drops the action entry once every class is empty. - `M._reset()` — test-only: clear active-map state. ## Usage ```luau local Zin = require("@builtin::modules.zinput") Zin.map.ensureActive() local eff = Zin.map.effective() print(eff.actions.jump.classes.touch ~= nil) -- synthesized or explicit -- Write the active scheme out as a control-asset map: Zin.map.bake("my_scheme") ```

# observe What became of a device event, and why a live control did not fire. Backs `Zin.observe`. A control can be live, its device can be producing input this frame, and the game can still receive nothing. The input layer works out why every frame — a gate said no, a live map stood the group down, the UI took the pointer, nothing subscribed — and until something reads that decision it is thrown away. This module keeps it. ## The one question ```luau local Zin = require("@builtin::modules.zinput") Zin.observe.whySilent("look") --> { -- name = "look", -- layer = "scheme", -- map = "default", -- reason = "gateRefused", -- means = "the control's own gate answered no. A look control gates on -- the player steering, so a cursor crossing the window reads -- nothing.", -- kind = "axis2", -- subscribers = 1, -- declaredClasses = { "kbm", "gamepad", "touch" }, -- carriedBy = { "scheme" }, -- } ``` One reason, from a closed set, with the particulars behind it. `Zin.observe.reasons()` lists the whole set with a sentence each, so what an answer means is readable from inside the engine rather than only from the source. ## The closed set, nearest cause first | Reason | What it says | |---|---| | `noMapActive` | No input map is activated at all, so there are no controls. | | `unknownControl` | No live map declares a control under that name. | | `groupSuppressed` | A live map stands this control's group down. | | `contextInactive` | The control declares a context that is not on top of the stack. | | `heldOnArrival` | A device the control binds was already held when the control went live. The control reads the hold as held; `pressed` and `released` wait for the next press. | | `gateErrored` | The control's own gate raised. | | `gateRefused` | The control's own gate answered no. | | `noBindingForDevice` | The control declares no binding for any device class this session has. | | `belowDeadzone` | The devices produced a reading and the control's own deadzone zeroed it. | | `noSubscriber` | The control is delivering and nothing subscribed to it. | | `atRest` | The control is live and its devices are not being driven. | | `delivering` | The control is delivering its value to its subscribers. | `atRest` and `unknownControl` are the two an outside reader could never tell apart: a value reader answers `0` for both. This is where they separate. ## Which layer answered Three naming layers coexist: the live maps' controls, and the `Zin.actions` and `Zin.axes` registries a control asset replaced. Their name sets are separate — `toolMove` is a live control and not an action, `move_forward` is an action and not a control — so a name that is live in one reads as silence in the others. Every answer carries `layer` (the one that answered) and `carriedBy` (every layer that has the name at all). ## One control, everything about it ```luau Zin.observe.control("move") --> { -- name = "move", live = true, carriedBy = { "scheme" }, -- entries = { { map = "default", kind = "axis2", subscribers = 1, -- gated = false, classes = { kbm = {...}, gamepad = {...} } } }, -- lastTick = { value = { x = 1, y = 0 }, active = true, delivered = true }, -- why = { reason = "delivering", ... }, -- } ``` ## The frame ```luau Zin.observe.frame() --> { frameId = 41207, window = "one tick of binding dispatch — the most -- recent one", context = "default", maps = {...}, controls = {...}, -- cost = { frameId = 41207, maps = 3, controls = 27, ms = 0.09 } } ``` The window is **one tick** — the most recent one — rather than a sum since anything last looked, and reading takes nothing away from the next reader. Two observers in the same frame both get the truth. `cost.ms` is that one tick's own milliseconds. The same number lands in `profiler.stats("*zin.scheme.advance*")` as `script.zin.scheme.advance`, where it carries the average and the peak across every tick as well. ## Reading it without running a script `/runtime/input` serves the same answers as a file, beside the engine's own account of the frame's device events: ``` /runtime/input the whole document /runtime/input/events the frame's device events, each with `consumed_by_ui` /runtime/input/device pointer lock, UI focus, the session's device classes /runtime/input/mapping this module's account of the tick /runtime/input/controls/<name> one control's entry from that account ``` `input.observe()` returns the same document to a script. Building this layer's half costs a frame of work, so it is built only while something is reading: a read of either surface arms it for a window of frames, and the document's `mappingStatus` says whether the half it carries is `current`, `stale`, `arming` (this read armed it — read again), or `unarmed`. `Zin.observe.armedFrames()` is how many frames that window has left; each frame that passes takes one off it, and the per-frame publish spends the window rather than renewing it, so it runs out once the last reader stops asking. A window that runs out — or that `Zin.observe.arm(false)` closes — drops the report the engine was carrying with it, so a frame nobody observes costs what it cost before anything ever read. `Zin.observe.arm(true)` holds it open for a tool that wants it continuously. ## Which frame an answer covers `Zin.observe.frame()` and `Zin.observe.control()` report the most recent tick of binding dispatch. `Zin.observe.whySilent()` resolves against the devices as they are at the moment of the call, so it answers for a control the tick has never reached and for a name that does not exist. The engine publishes its own half at the end of a frame, after the frame has been drawn and before the frame's events are cleared. A read taken from a script during frame N therefore answers for frame N-1.

# pointer Pointer-lock mutation API. Wraps the internal `__zero_input.requestPointerLock` / `releasePointerLock` primitives and backs `Zin.pointer`. The read side lives on `Zin.state.pointerLocked()`; this module owns the write side plus a convenience read alias. ## Exports - `M.lock()` — request pointer lock (cursor grab + hide). Native takes effect immediately; WASM queues for the next user gesture. - `M.unlock()` — release pointer lock (cursor ungrab + show). - `M.locked() -> boolean` — convenience read alias, identical to `Zin.state.pointerLocked()`. ## Usage ```luau local Pointer = require("@builtin::modules.zinput.pointer") Pointer.lock() if Pointer.locked() then -- pointer captured end Pointer.unlock() ``` ## Notes - Native: `lock()` takes effect immediately. WASM: queued for the next user gesture (browser policy). Both routes share the same underlying FFI mutation; the platform enforces the difference. - Write side and read side are intentionally split: `Zin.state.pointerLocked()` stays alongside the rest of the per-frame snapshot accessors; mutation lives here for discoverability. - All entry points are no-ops when the `__zero_input` FFI is absent (off-host script execution).

# profile Binding profile registry — bundles of named actions / axes / chords that can be loaded from `@builtin::profiles.<name>` Luau modules or saved as JSON under `/zero/profiles/<name>.json`. Activating a profile reconfigures `Zin.actions` / `Zin.axes` / `Zin.chords` in one shot. The active profile is per-client session state held in memory — each session resolves it fresh from the device-appropriate default (or an explicit `activate`), so it is never written to the world source tree. This is the persistence and activation layer for those three registries; `Zin.map.activate` applies a materialized map through it. A scheme is authored as an `.inputMap` of `.inputBinding` controls and goes live through `Zin.scheme` — `man topics/input` covers that path. ## Exports - `M.register(name: string, profile: any) -> (boolean, string?)` — validate + register a profile descriptor. - `M.load(name: string) -> (boolean, any?)` — load + register a profile from `@builtin::profiles.<name>` or `/zero/profiles/<name>.json`. - `M.activate(name: string, opts: ActivateOpts?) -> (boolean, string?)` — apply a profile's sections + set the active name for this session. - `M.current() -> string?` — active profile name for this session, or nil. - `M.get(name: string?) -> any?` — descriptor for the active profile or the named profile. - `M.ensureActive(fallback: string?) -> (boolean, string?)` — bootstrap helper for controllers. - `M.save(name: string, opts: SaveOpts?) -> (boolean, string?)` — capture current Zin.actions / Zin.axes / Zin.chords to JSON. - `M.delete(name: string) -> (boolean, string?)` — remove a user profile's JSON file (built-ins refused). - `M.export(name: string, vfsPath: string) -> (boolean, string?)` — write a serializable JSON snapshot to any VFS path. - `M.import(vfsPath: string, opts: ImportOpts?) -> (boolean, string)` — read + register a profile JSON from any VFS path. - `M.list() -> { string }` — sorted union of built-in + user profile names. - `M.onChange(cb: ChangeCallback) -> number` — subscribe to activation / re-register events; returns a handle. - `M.off(handle: number)` — drop a subscription. - `M._reset()` — test-only: clear in-memory state. Types: - `Profile = { name, description?, actions?, axes?, chords?, intentional_pairs? }` - `ActivateOpts = { reactivate: boolean? }` - `SaveOpts = { description: string?, overwrite: boolean? }` - `ImportOpts = { overrideName: string?, activate: boolean? }` - `ChangeCallback = (string, any) -> ()` ## Usage ```luau local Profile = require("@builtin::modules.zinput.profile") Profile.register("my-bindings", { name = "my-bindings", actions = { jump = { context = "default", bindings = { { kind = "key", code = "Space" } } } }, }) Profile.activate("my-bindings") local h = Profile.onChange(function(name) print("active:", name) end) Profile.off(h) ``` ## Notes - Built-in profiles ship as Luau modules at `@builtin::profiles.<name>` and can include functions (gate, curve). User profiles are pure JSON at `/source/profiles/<name>.json` — function fields are dropped on save. - Activation is no-op when the requested name is already active unless `opts.reactivate = true`. - `save` / `export` VFS writes use the bare `/source/...` mount — the `/zero/...` prefix is read-only and silently drops writes. - The conflicts module's `_setPersistFn` hook is wired here so mark/unmark of intentional pairs persists into the active user profile automatically.

# quiesce The dirty signal `Zin.tick`'s quiescence gate reads. A mutation of the mapping layer's structure outside the tick itself — a map activation, a context change, a virtual-control write, a forced surface class, a new axis over an already-held key — calls `mark()`, and the tick runs a full pass while a mark is pending. A tick with no pending mark and no live input events skips its pass. `mark()` raises the signal, `pending()` answers whether a full pass is owed, and the generation pair (`generation()` / `ack(g)`) keeps a mark raised DURING a full pass pending for the next one, so a mutation racing a tick is never lost. Internal to the zinput package: the mapping-layer modules mark it, the tick drains it.

# rebind Interactive binding-capture sessions. A session listens for the next keyboard or mouse event and produces a binding descriptor; the host can then `apply()` it back into a `Zin.actions` action or a `Zin.axes` axis. Backs `Zin.rebind`. Drives the editor's Input tab rebinding UI and is usable from any Luau code. Modifier-aware (captures `modKey` when Ctrl/Shift/Alt is held) and timeout-bounded (default 30 s). ## Exports - `M.begin(opts: BeginOpts?) -> Session` — start a capture session; subscribes to `Zin.events.on` and returns a session table. - `M.cancelAll()` — cancel every live session. - `M.liveCount() -> number` — number of capturing sessions still active. - `M._advanceTime(dt: number)` — internal: per-frame timeout sweep wired by `Zin.tick`. Types: - `Session = { target, name, slot, mode, status, timeoutSec, result, consume, apply, cancel, conflicts, ... }` - `Owner = { target: string, name: string, slot: any }` - `BeginOpts = { target?, name, slot?, mode?, timeoutSec?, filter?, applyOnCommit?, onCommit?, ... }` ## Usage ```luau local Rebind = require("@builtin::modules.zinput.rebind") local s = Rebind.begin({ target = "action", name = "jump", timeoutSec = 10, applyOnCommit = true, onCommit = function(session) print("captured:", session.result()) end, }) -- The session subscribes itself; just wait for the next tick. if s.status == "capturing" then -- ... end ``` ## Notes - Press Escape during a capture to cancel; a pure modifier press is ignored (the session waits for the base key). - A session that captures with Ctrl/Shift/Alt held returns a `modKey` binding rather than a plain `key` binding. - `session.conflicts()` is self-filtered — rebinding an action to a key it already owns reports no conflict. - Hot-reload safety: the live-sessions list survives require-cache reload; the `_advanceTime` wiring from `zinput/init.luau` keeps timeouts firing across re-requires. - `applyOnCommit = true` calls `apply()` automatically after commit; errors during apply are still visible via `result()` and `conflicts()`.

# scheme The live set of activated input maps, and the per-frame dispatch that turns their bindings into the events a controller subscribed to. Backs `Zin.scheme`. ## Nothing is live until something activates a map A controller holds an `inputMap` ref, activates it when it takes control, and releases it when it gives control up. The live binding set is exactly the union of what current holders asked for, so an empty world has an empty input surface and an empty screen — no scheme is active by default, and nothing appears on a phone that no component consumes. For a component that drives the thing it lives on, taking control and being awake are the same moment, so `awake` / `onDestroy` are the hooks. For anything a player enters and leaves — a car, a turret, a menu — they are not: the car is still there after the player gets out, so the map goes live on entry and stands down on exit. ```luau public = { inputMap = Field.assetRef("inputMap", "@builtin::inputMaps.default", NoSync, "The controls this component reads."), speed = Field.number(5, NoSync, "World units per second at full deflection."), jumps = Field.number(0, NoSync, "How many times the jump control was pressed."), } local controls function awake() controls = public.inputMap:activate() controls.move:onInput(function(v, dt, active) if not active then return end local p = public.entity.position local step = public.speed * dt public.entity.position = { p.x + v.x * step, p.y, p.z - v.y * step } end) controls.jump:onPressed(function() public.jumps += 1 end) end function onDestroy() public.inputMap:deactivate(controls) end ``` What `activate` returned is this holder's **claim** on the map. Keep it and hand it back: releasing the claim disconnects the subscriptions made through it, and the map stays live for whoever else holds it. A claim also ends with the context that opened it. A component's claim is released when that component instance goes, whether or not it writes an `onDestroy`: the component is removed from its entity, the entity despawns, or the component hot-reloads (its new life's `awake` activates again). A claim a module body opens ends at the module's next run. `scopes.list()` names each standing claim as a `luau` row under the context it follows, and `scopes.release(tag)` ends it with everything else that context holds. ## Several maps compose by all being live The player's surface is the union of every live map, which is what a game needs: picking up a gun activates a weapon map and Fire appears beside the movement stick; dropping it takes away Fire and nothing else. Two controllers that each activate a map with a movement stick produce two sticks — visible, attributable to the second controller, and the importer's to resolve. Two components activating the *same* map read one set of controls, each through its own claim. A claim's release disconnects what that holder subscribed and leaves the other holders' handlers attached; the map leaves the live set when its last claim goes. `Zin.scheme.live()` reports each live map's `holders` — how many claims are still standing on it — and `heldBy`, which says who each of them is: `kind` (`"component"` or `"execute"`), `entity`, `source` as the `chunk:line` the activation was opened at, and the `label` the caller passed to `activate("weapon-swap")`. ## An activation reads the map's children as they stand now Every `activate` re-reads the map's `<name>.inputBinding/` children, so a control added, edited or removed since the map came up reaches the running set on the next activation. The controls are edited in place, so it reaches the holders already standing on the map too: their handles keep working and every subscription they made carries on across the edit. ## A map can stand another one down Composition cannot say "not while this is happening". A map declares the `group` its controls belong to and the groups it `suppresses` while live: ```luau -- driving.inputMap/init.luau return { group = "vehicle", suppresses = { "player" } } ``` Every live map in a suppressed group reads nothing and draws nothing for as long as the suppressor is live. It is **not** released: it keeps its holders, its handles and every subscription, and resumes untouched when the suppressor goes. So a car stands down walking without ever finding, owning or releasing the walk map — and a menu stands down the whole game with one declaration. A map that declares no `group` belongs to one named after itself, so it collides with nothing and can still be named by something that wants to stand it down. Declaring neither field composes with everything. The drop frame runs on the tick a map stands down, so a stick held at that moment reports its neutral value once and whatever it drove stops rather than sticking. ## Dispatch `_advance(dt)` evaluates every live binding once per tick, across every device class it carries — a tablet with a keyboard attached and a desktop with a pad plugged in both answer to whichever the player actually used, rather than to a nominated primary. | Event | Fires | Cadence | |---|---|---| | `input` | Once while the binding is active with `(value, dt, true)`, and once more on the frame it goes inactive with the neutral value and `false` | Per engine frame | | `pressed` / `released` | A button's edges | Per engine frame | | `changed` | When the value differs from the previous tick | Per tick | `input` firing on the drop frame is what lets one handler both start and stop the motion it drives, without a second subscription to notice the release. A frame carries more than one tick whenever a host calls `Zin.tick` beside the tick the engine already runs, or fast-forwards several ticks to converge smoothing. Evaluation follows the tick, so the smoothing lands and `changed` reports each step of it. `input`, `pressed` and `released` are stamped with the engine frame they were reported for, so the frame carries one of each however many ticks ran in it. Both edge sources hold this frame's edge for as long as the frame lasts: the input snapshot's `pressed` flag, and `Zin.virtual`'s previous-frame button table. That is what a second reporting pass reads as another press. `Zin.scheme.fired`'s `count` is the number of engine frames a control was active since it last cleared, so a button held across two frames reads 2 for one press. The `pressed` / `released` subscriptions are what count edges. A binding that declares a `context` is live only while that context is on top, the same activity rule actions apply. An `axis1` / `axis2` control's own `deadzone`, `curve`, `invert` and `smoothing` are applied here, in that order, to the reading its classes produced — after `as` / `unitsPerSecond` / `scale` have brought every class into the control's unit, so one deadzone written in that unit covers the stick, the mouse and the finger together. A deadzone is per value on an `axis1` and **radial** on an `axis2`. The shaping maths is `Zin.utils`', shared with the axis registry. `smoothing` is the one piece of the four that is stateful: the value is held per control across frames and approaches whatever the devices report next over the declared time constant. A control that leaves its context, whose group stands down, or whose gate closes has a neutral target, so it drains rather than cutting — and reports its drop frame when it arrives. ## Validation happens before anything registers Activation reads every child binding, collects every fault across all of them, and raises once with the lot. A map with three broken bindings names all three rather than one per attempt, and a map that fails validation registers nothing — a half-live map is worse to debug than one that refused to come up. ## Exports - `M.activate(mapRef, label?) -> { [name]: Handle }` — activate a map, return its handles. `label` names this activation in `M.live()`. - `M.deactivate(mapRef) -> boolean` — release a claim on a map. - `M.live() -> { { guid, name, group, suppresses, suppressedBy, holders, heldBy, bindings } }` — every live map in activation order. What tells an author why two sticks are on screen, or why the controls they activated answer to nothing, or who is holding the map standing them down. - `M.bindings() -> { { map, group, name, label, kind, context, suppressedBy, subscribers, record } }` — every live binding in activation order. The touch overlay builds from this, so what is on screen is exactly what some awake component asked for and nothing is standing down. - `M.suppressed() -> { [group]: mapName }` — which groups are standing down and what put each one down. - `M.has(name) -> boolean` — whether a named control is live anywhere. - `M._advance(dt)` — one tick of dispatch; wired into `Zin.tick`. - `M._childBindings(mapRef) -> { ref }` — the map's `.inputBinding` children, in name order. - `M._reset()` — test-only: drop every live map without firing anything. Types: - `Handle = { onInput, onPressed, onReleased, onChanged, label, kind }` — a thin view over the binding asset's own event facade, so a subscription made through a handle is the same one `ref.events.input:connect` would make and ends the same way.

# state Synchronous per-frame input state. Thin wrapper over the internal `__zero_input.snapshot()` FFI; backs `Zin.state`. Pointer-lock mutations live on `Zin.pointer`; the read-side `pointerLocked()` query lives here with the rest of the snapshot accessors. Held-time and synthetic-repeat state is owned by this module and updated via `_observeEvent` from the `Zin.tick` coordinator. ## Exports - `M.get() -> Snapshot` — full per-frame snapshot table; empty when the FFI is unavailable. - `M.keyDown(key: string) -> boolean` — held this frame. - `M.keyPressed(key: string) -> boolean` — edge-pressed this frame. - `M.keyReleased(key: string) -> boolean` — edge-released this frame. - `M.mousePosition() -> (number, number)` — current mouse `(x, y)`. - `M.mouseDelta() -> (number, number)` — `(dx, dy)` since last frame. - `M.scrollDelta() -> number` — wheel delta this frame. - `M.mouseButtonDown(idx: number?) -> boolean` — held; default idx = 0. - `M.mouseButtonPressed(idx: number?) -> boolean` — edge-pressed. - `M.mouseButtonReleased(idx: number?) -> boolean` — edge-released. - `M.pointerLocked() -> boolean` — pointer-lock query (writes are on `Zin.pointer`). - `M.uiWantsPointer() -> boolean` — `true` when the UI layer claims pointer focus. - `M.uiTookPress() -> boolean?`: `true` while a UI widget sensing a click or a drag holds the pointer press in progress; `false` with no press held and for a press on empty space; `nil` from the moment a press arrives until a UI pass has decided it. - `M.setRepeatDefaults(opts: RepeatOpts?)` — set default delay / period for synthetic repeat. - `M.getRepeatDefaults() -> RepeatDefaults` — read the active synthetic-repeat defaults. - `M.keyHeldTime(code: string) -> number?` — seconds since the key was pressed (or nil). - `M.mouseButtonHeldTime(idx: number?) -> number?` — seconds since a mouse button was pressed. - `M.keyRepeatFired(code: string, opts: RepeatOpts?) -> boolean` — should a synthetic key-repeat fire this frame? - `M._observeEvent(ev: any)` — internal: per-event observer, wired from the tick coordinator. - `M._reset()` — test-only: clear held-time and synthetic-repeat state. Types: - `Snapshot = { [string]: any }` - `RepeatOpts = { delay: number?, period: number? }` - `RepeatDefaults = { delay: number, period: number }` ## Usage ```luau local State = require("@builtin::modules.zinput.state") -- Rebind-capture-style probe: how long has a key been held, and is a -- synthetic repeat due this frame? local heldFor = State.keyHeldTime("Space") if heldFor and State.keyRepeatFired("Space", { delay = 0.4, period = 0.08 }) then -- charge-attack tick, key-repeat menu navigation, etc. end ``` ## Notes - Mouse-button indices are 0-based and follow `InputResource.mouse_buttons[]` (0=left, 1=right, 2=middle). Winit's `0=L, 1=M, 2=R` is converted in Rust before reaching Luau. - `mousePosition` / `mouseDelta` return two numbers, not a table. - Held-time / repeat state is updated by the `Zin.tick` coordinator via `_observeEvent` — without a tick, those queries return `nil` / `false`. - `keyRepeatFired` advances state on every call — call once per frame per consumer. - All entry points are no-ops or return zero values when `__zero_input` is missing (off-host execution). - Gameplay subscribes to controls through an activated `.inputMap`; this module is the raw snapshot layer those bindings evaluate against, used directly by rebind capture, the emulation floor, and dev tooling. Polling it from a world makes that world keyboard-only and draws the `engine-raw-key-poll` diagnostic at author time.

# surface Active input-surface classification. Answers "what device class is the player using right now" — `"touch"` or `"kbm"` — and notifies on change; backs `Zin.surface`. Virtual touch controls mount whenever the surface is `"touch"`, in edit and play alike. ## The frame rule Resolution runs once per engine frame from the `Zin.tick` pipeline: a per-event observer collects which device classes produced events this frame, and the frame resolver applies one rule — **touch wins a frame that contains any touch event**. The primary touch contact also drives the mouse path (a projected `mouse.down`/`move`/`up` in the same frame), so classifying by the latest single event would flip to `kbm` on every tap; the frame rule keeps projection-generated mouse events from masking the touch that caused them. ## Capability default Before any input arrives, the class defaults to `"touch"` on sessions with a touch source (`touch_capable` — latched by the first contact or declared at boot by Android / the web client) and `"kbm"` otherwise, so touch-first devices present touch controls from the first frame. ## Exports - `M.current() -> string` — the active surface class, `"touch"` or `"kbm"`. - `M.onChange(fn: (string, string?) -> ()) -> number` — subscribe to class flips; `fn(now, prev)` where `prev` is the previously reported class (the first-tick capability default when no event had resolved yet). Returns a handle. - `M.off(handle: number) -> boolean` — unsubscribe an `onChange` handle. - `M._beginFrame()` — internal: clear this frame's per-class flags; wired by `Zin.tick`. - `M._observeEvent(ev: any)` — internal: per-event observer; wired by `Zin.tick`. - `M._resolveFrame()` — internal: resolve this frame's class and fire `onChange` on a flip; wired by `Zin.tick`. - `M._reset()` — test-only: clear resolution state and subscriptions. Types: - `SurfaceClass = "touch" | "kbm"` ## Usage ```luau local Surface = require("@builtin::modules.zinput.surface") if Surface.current() == "touch" then showTouchOverlay() end Surface.onChange(function(now, prev) print("surface flipped to", now) end) ``` ## Notes - `onChange` fires from the `Zin.tick` pipeline on the frame the class flips; without a tick, resolution never advances and `current()` keeps returning the capability default. - The change signal compares against the last reported class, whose initial value is the capability default captured on the first tick of the session — a phone's first contact confirms the default silently, while a desktop touchscreen's first contact reports kbm→touch. - Handler errors during `onChange` dispatch are caught and logged, not propagated.

# test Test-input helpers that simulate keyboard / mouse / scroll events and yield one engine frame so the queued events drain into the next snapshot. Backs `Zin.test`. Helpers do NOT call `Zin.tick()` — explicit dispatch control is left to the caller. ## Exports - `M.runFrame()` — yield one engine frame so simulated events drain. - `M.pressKey(code: string)` — hold the key DOWN and leave it held. A second `pressKey` on a held key produces no new press edge — pair every `pressKey` with a `releaseKey`, or use `tapKey` for a full press. - `M.releaseKey(code: string)` — simulate a key release. - `M.tapKey(code: string, duration: number?)` — press + optional hold + release. The helper for "the player taps a key". - `M.pressMouse(button: (number | string)?)` — simulate a mouse button press (default 0 = left). The index and the name (`"left"`/`"right"`/`"middle"`) reach the same button. - `M.releaseMouse(button: (number | string)?)` — simulate a mouse button release. - `M.clickMouse(button: (number | string)?, x: number?, y: number?, duration: number?)` — full click, optionally moving the cursor to `(x, y)` first and holding for `duration` seconds. - `M.moveMouse(x: number, y: number)` — set absolute cursor position. - `M.moveMouseBy(dx: number, dy: number, steps: number?)` — move the pointer by a relative motion. The motion lands in this frame's `mouse_delta` and the position advances by the same amount, so the same `(dx, dy)` repeated keeps producing look movement. `steps` spreads it over that many frames, each carrying its share; 1 when omitted. - `M.scroll(dy: number)` — simulate a wheel scroll. ## Usage ```luau local Test = require("@builtin::modules.zinput.test") Test.tapKey("KeyM") -- press + release Zin.tick(1 / 60) -- explicit dispatch assert(#Zin.scheme.fired() > 0) Test.pressKey("Space") -- key goes DOWN and stays held... Zin.tick(1 / 60) Test.releaseKey("Space") -- ...until released Test.clickMouse(0, 100, 200) ``` ## Notes - Each helper queues the simulated event and yields one frame (`task.wait(0)`) so the snapshot reflects the input on return. - Simulated key state persists like real key state: after `pressKey` the key stays held until `releaseKey`, and only the release re-arms the next press edge. `tapKey` is the edge-per-call helper. - Helpers do NOT call `Zin.tick()` — a test that relies on a control's handlers, or on `Zin.input.on*` / `Zin.actions.bind`, must run a tick itself. `Zin.scheme.fired()` is what turns a simulated input into an assertion: it reports which controls fired since it last read, and from which device class. - Mouse button index is 0-based (0=left, 1=right, 2=middle) following the `InputResource.mouse_buttons` convention. - All helpers are no-ops when the `__zero_input.simulate*` FFI is missing (off-host execution).

# text Requiring this module raises a fatal error: `zinput.text` is not available in this build. ## Exports None — the module raises on `require`.

# touch Per-finger touch contacts. Thin wrapper over the `touches` array of the internal `__zero_input.snapshot()` FFI plus pre-baked `touch.*` event listeners; backs `Zin.touch`. ## Exports - `M.slots() -> { TouchPoint }` — all active contacts this frame, in begin order. - `M.count() -> number` — number of active contacts this frame. - `M.byId(id: number) -> TouchPoint?` — look up an active contact by finger id. - `M.primary() -> TouchPoint?` — the primary contact (slot 0), if a finger is down. - `M.capable() -> boolean` — whether the session has (or has declared) a touch source. - `M.onBegan(fn, opts?) -> handle` — subscribe to `touch.down` events. - `M.onMoved(fn, opts?) -> handle` — subscribe to `touch.move` events. - `M.onEnded(fn, opts?) -> handle` — subscribe to `touch.up` events. - `M.onCancelled(fn, opts?) -> handle` — subscribe to `touch.cancel` events. - `M.off(handle)` — unsubscribe a handle returned by any `on*` function. Types: - `TouchPoint = { id: number, slot: number, x: number, y: number, dx: number, dy: number, pressure: number, phase: string, pointer: boolean }` ## Contact record fields | Field | Meaning | |---|---| | `id` | The platform's stable finger id for the contact. | | `slot` | Lowest index free when the contact began; 0 = primary (also drives the mouse path). | | `x`, `y` | Screen position. | | `dx`, `dy` | This frame's accumulated delta. | | `pressure` | Normalized 0..1 (1 when the platform reports none). | | `phase` | `"began"` \| `"moved"` \| `"stationary"` \| `"ended"` \| `"cancelled"` (ended/cancelled contacts survive exactly one frame). | | `pointer` | Whether the contact drives the UI's pointer, so a widget under it can take its press. The primary contact does, unless it began on a region an on-screen control holds (`input.setTouchClaims`) or is a simulated contact for the game alone. | ## Usage ```luau local Touch = require("@builtin::modules.zinput.touch") for _, t in ipairs(Touch.slots()) do print(t.slot, t.x, t.y) end local h = Touch.onBegan(function(ev, gpe) print(ev.id, ev.x, ev.y) end) ``` ## Notes - `onBegan` / `onMoved` / `onEnded` / `onCancelled` are pre-baked kind filters over `Zin.events.on` — handlers dispatch from the `Zin.tick` pipeline and receive `(ev, gpe)` where `gpe` is the UI-focus flag at fire time. Return `"sink"` to stop lower-priority handlers from seeing the same event. - All entry points return empty / falsy values when `__zero_input` is missing (off-host execution).

# touchControls On-screen touch overlay: a renderer + worker, not an event-driven widget net. The worker reads raw touches (`Zin.touch.slots()`) and routes them into zone geometry derived from the live bindings, writing `Zin.virtual` (stick/button/drag). `Zin.tick` runs the worker early — after `Virtual._beginFrame()` clears the previous tick's drag deltas and before the binding evaluators read `Zin.virtual` — so the touch binding kinds consume those values on the SAME tick they're written. The renderer (a `canvas` screen declaring `pointerEvents = "none"`) only draws and takes no pointer, so every widget beneath it, in any screen at any layer, still receives the presses that land on it. Backs `Zin.touchControls`. ## Who a contact belongs to - A contact that begins on a drawn button, the fan or an open sheet entry is that control's alone. The overlay states those circles to the input layer with `input.setTouchClaims` (in the window pixels contacts arrive in), and a contact beginning inside one drives no pointer: no widget beneath the button is pressed by it, and no mouse button reads held. - A contact that begins anywhere else drives the pointer as the primary contact always does, and the UI decides whether a widget takes its press. When one does, the contact is the widget's: on the tick after it began, the stick or look zone that took it lets it go before it has moved anything. - Everything left falls to the stick and look zones below. ## Coordinate space Touch contacts (`Zin.touch.slots()` — `t.x`/`t.y`/`t.dx`/`t.dy`) carry PHYSICAL framebuffer pixels — `InputResource` keeps the raw platform coordinates unconverted, the same space `getViewportSize()` reports. Every widget geometry computation — button/stick positions, the canvas root's `style.width/height` — lives in `ui.screenSize()`'s LOGICAL space (the space `ui.registerScreen` / `ui.getLayoutInfo` use), which differs from the physical space whenever the display's `pixels_per_point ~= 1` (HiDPI). An internal `logicalScale` helper computes the `screenSize / viewportSize` ratio once per tick, and every touch coordinate this module reads is scaled by it before use. ## Safe area Every control keeps inside the part of the overlay no system bar, display cutout or on-screen keyboard covers. `ui.safeArea()` states that part of the picture the overlay's screen lays out in (the scene viewport's, at `ui.screenSize()`), and the overlay takes it into its own space, which leaves a safe rect `L`/`T` in from the left and top edges, `SW` x `SH` in size (the whole picture where the picture lies inside the window's safe area, as a Scene panel does). The zones, the button column and the stick all lie inside it, a contact beginning outside it starts none of them, and `layout().safe` reports it. ## Size on the glass Every radius, inset and gap is in the points `ui.screenSize()` reports. On a phone or a tablet a point covers at least the display's density (`ui.displayScale()`), so the controls keep one physical size on every phone and in both orientations: a 40 point "small" button is 80 dp across, past the 48 dp touch target. `layout()` reports the radii in points; multiply by the pixels one of the overlay's points covers for their size on the glass: the width `ui.getLayoutInfo("zin_touch_controls_canvas")` gives its canvas, over `layout().safe.width`, times `ui.pixelRatio()`. Button labels are canvas text sized from the button's radius, so the user's text scale leaves them inside their circles. ## Zone geometry (fractions of the safe rect, `SW`/`SH`) - **stick** - activation region `x < L + 0.4*SW and y > T + 0.4*SH` (left 40%, bottom 60%). A floating stick: the base spawns at the touch's start point, the thumb clamps to a 96px radius, and the vector fed to `Zin.virtual.setStick` is `offset / 96` with `y` screen-down positive (matching `setStick`'s documented convention). - **buttons** — one circle per `touchButton` binding among the LIVE bindings (`Zin.scheme.bindings()` — what awake components asked for, so every button corresponds to a control some world actually consumes), budgeted into a two-column band along the safe rect's right edge, between 45% of its height and its bottom edge less the inset. Buttons sort by `(priority, id)` - the touchButton's `priority` option, authored default 50, synthesized default 100 — with same-`group` buttons kept adjacent, then fill the two columns growing inward from the bottom-right corner. Each circle is the radius ITS OWN `size` option names (`small` 40px, `medium` 56px, `large` 72px), so a scheme mixing a `large` action button with a `small` pause pip draws them at 72 and 40 whatever else is live; the largest of those radii is the slot pitch and the row budget the column packs on, so the set decides where a button sits and the button's own declaration decides how big it is. More than 4 authored buttons walk every button one step down its own size, which `layout()` reports as `appliedSize`. Anything past the two-column budget doesn't render individually — the last slot becomes a `...` fan button; tapping it opens a rows-of-3 grid sheet holding the overflow (same tap/hold behavior as a regular button), and tapping the fan again, or any layout-invalidating context/map change, closes it. - **drag** - everything else in the safe rect at `x >= L + 0.4*SW` (accumulates into `Zin.virtual.addDrag` every tick from the contact's per-frame delta). - the remaining top-left quadrant (`x < L + 0.4*SW and y <= T + 0.4*SH`) is unclaimed - free for other UI. A control is drawn only when it is live AND subscribed: a button offered for a control nothing listens to is an offer that cannot be kept, so the layout tracks `Zin.scheme.subscriberEpoch()` alongside `Zin.scheme.generation()`. Only bindings whose `context` matches the current input context (`Zin.context.current()`) contribute controls, so a vehicle-context Brake button neither renders nor claims touches while the default context is on top. The layout rebuilds when the live set, the subscriber count, the context, the screen size or the safe area changes. A record-shaped map applied through `Zin.map.activate` contributes its touch classes too. A control whose name is already live as an asset is skipped, so a world carrying both never draws the same control twice under two ids. A contact is assigned to a region on its `began` phase only (a contact-id -> assignment map) and released on `ended` / `cancelled` / vanishing from `Zin.touch.slots()`. Button circle hit-tests (visible buttons, the fan, the open sheet's entries) take precedence over the stick/drag zone checks — a button is a precise target and wins wherever it lands, including inside a zone (the open sheet grows leftward and can reach the stick zone on a narrow screen). A contact claimed by the stick or a button never also feeds the drag zone (a contact is assigned to exactly one region), so its projected-mouse / `touchDrag` state never competes with a look/aim consumer reading `Zin.virtual.drag` — no extra suppression is needed beyond the region assignment itself. A pinch and the drag region share the same space, so the two are arbitrated once per tick. The contacts the stick and the buttons hold are published to `Zin.gestures`, which forms its two-finger pair out of what is left; while the active map declares a `touchPinch` control and that pair is down, the drag region reports nothing. A pinch made over the look surface therefore reaches the pinch control alone, a thumb resting on the movement stick is no part of a pinch so the other thumb keeps looking, and lifting one finger from a pinch drags again. ## Rendering One screen (`"zin_touch_controls"`, layer 200) whose root is a single transparent `canvas` sized to the safe rect, taking no pointer. The screen lays out inside the safe area, so the canvas starts at the rect's corner and every command is drawn from there. Commands rebuild only when the draw state actually changed (a dirty flag) — buttons are always drawn while mounted; the stick draws only while a stick contact is live. ## Auto-mount Every `_advance()` checks `Zin.surface.current()` — `"touch"` registers (lazily, once) and shows the screen; anything else hides it. Visibility is purely the surface class, in edit and play alike. `M._forceMount(true)` bypasses the surface gate for headless testing. Touch routing itself is unconditional: it runs every tick regardless of mount state, so a forced or real touch always drives `Zin.virtual` even before the overlay is visible. ## Adding a control while the world runs The overlay draws a button for a control that is live and subscribed, so adding one at runtime is adding a control asset, activating it, and subscribing to it — the three steps a component takes in `awake`, in one call: ```luau local h = Zin.touchControls.addControl({ name = "cast", label = "Cast", kbm = Zin.bindings.key("KeyF"), gamepad = Zin.bindings.padButton("north"), onPressed = function() castSpell() end, }) Zin.touchControls.removeControl(h) ``` The control becomes a real `<name>.inputMap/<name>.inputBinding/` under `/source/runtimeControls/`, so it is inspectable, rebindable and publishable like any other, and its name is its id in `Zin.scheme.bindings()` / `Zin.scheme.fired()`. `removeControl` releases it — off the screen, out of the live set, every subscription dropped — and leaves the asset, so re-adding the same name reuses it and its guid. `M.button(opts)` is the same idea against a record-shaped map applied through `Zin.map.activate`, in two flavors: - `{ action = "jump" }` attaches an extra touchButton binding to an existing action on the active map. - `{ emit = "KeyF", label = "Cast" }` registers an overlay-only button: a synthetic `emit:KeyF` action carrying both the kbm key binding and the touchButton, so the emulation floor drives `KeyF` through the same binding/emulation path every other control uses. Both return a handle for `M.removeButton(handle)`. ## Exports - `M.addControl(opts: ControlOpts) -> number` / `M.removeControl(handle) -> boolean` — add or release a control while the world runs. - `M.controls() -> { { handle, name, path } }` — every control `addControl` currently holds, in the order they were added. - `M.button(opts) -> number` / `M.removeButton(handle) -> boolean` — the button API described above. - `M.claimed() -> { number }` — the contact ids currently claimed by the stick, a button, or the drag zone. - `M.primaryClaimed() -> boolean` — whether an on-screen control owns the primary contact. `Zin.bindings` consults this so a thumb on the stick or the look area does not also fire whatever the world bound to left-click. - `M.layout() -> { buttons, fan?, stick?, drag?, pinch?, width, height, safe }` - the current control layout (a copy): `buttons` is the budgeted, VISIBLE stack in draw order, each entry `{ id, label, actionName, priority, size, appliedSize, group, cx, cy, radius }` in `ui.screenSize()` space — `size` is the step the binding declared and `appliedSize` the one `radius` came from, which differ only where a rich scheme's step-down moved a button one size down; `fan` is nil unless the button set exceeded the two-column budget, else `{ open, cx, cy, radius, buttons }` — `open` is whether the overflow sheet is expanded, `buttons` the overflowed entries at their sheet positions (tappable only while `open`); `stick` / `drag` name the axis + virtual zone each region feeds, and `pinch` the axis the two-finger recognizer feeds; `safe` is the safe rect `{ x, y, width, height }` every control keeps inside. Backs custom control-placement UI and headless tests that aim simulated touches at a button's (or the fan's) center. - `M._advance()` — internal: one tick (touch routing, layout, auto-mount, repaint). Wired into `Zin.tick`, before `Zin.emulation._advance()`. - `M._forceMount(on: boolean)` — test-only: force mounting regardless of `Zin.surface.current()`. - `M._setViewportOverride(w?, h?)` — test-only: override the logical screen size the layout budget/fan math reads, without resizing the real window. Pass nil/nil to clear. - `M._reset()` — test-only: unmount, clear routing/layout state, drop buttons, release every runtime control, close the fan, and clear the viewport override.

# utils Pure helpers for the `Zin` input library. Mouse-button name↔index conversion, key-code normalization, held-modifier matching against a frame snapshot, and the shaping a reading passes through on its way to a consumer. No FFI; safe to call from any context. Stateless. ## Exports - `M.buttonName(idx: number) -> string?` — convert a 0-based mouse button index to its name. - `M.buttonIndex(name: string) -> number?` — convert a mouse button name to its 0-based index. - `M.normalizeKey(code: string) -> string` — canonicalize a key code: a single ASCII letter becomes its `Key<L>` code, a single digit its `Digit<N>` code; multi-character codes pass through. - `M.matchModifiers(snapshot: Snapshot, mods: Modifiers) -> boolean` — test whether the snapshot's held keys match the requested modifier set. - `M.applyCurve(curve, x: number) -> number` — apply a response curve to a reading. - `M.applyDeadzoneScalar(x: number, deadzone: number?) -> number` — 0 below the threshold, the reading at or above it. - `M.applyDeadzoneVector(v: Vector2, deadzone: number?) -> Vector2` — the same, measured **radially** on the magnitude of the pair. - `M.shapeScalar(x: number, deadzone: number?, curve, invert: boolean?) -> number` — deadzone, then curve, then inversion. - `M.shapeVector(v: Vector2, deadzone: number?, curve, invert: boolean?) -> Vector2` — the same for a pair. - `M.smoothToward(current: number, target: number, dt: number, tau: number) -> number` — one step of an exponential approach with time constant `tau`. A `curve` is `"linear"`, `"quadratic"`, `"cubic"`, a function of the reading, or `nil` for linear. Types: - `Snapshot = { [string]: any }` — the raw frame-snapshot table returned by `Zin.state.get()`. - `Modifiers = { ctrl: boolean?, shift: boolean?, alt: boolean? }` — modifier subset to require in `matchModifiers`. - `Vector2 = { x: number, y: number }` — a two-component reading. ## Shaping The order is deadzone, curve, invert, then the approach toward the shaped target. A scalar takes its deadzone per value; a pair takes it radially, so a diagonal is not clipped into a cross by two independent thresholds. `"quadratic"` squares while keeping the sign, so a half-pushed stick reads a quarter in the direction it was pushed. A curve function that raises, or answers with anything other than a number, leaves the reading as it was. This is the one implementation of that shaping: `Zin.axes` reads it for a registered axis and `Zin.scheme` for an `axis1` / `axis2` control, so an axis and a control declaring the same numbers read the same. ## Usage ```luau local Utils = require("@builtin::modules.zinput.utils") local name = Utils.buttonName(0) -- "left" local idx = Utils.buttonIndex("right") -- 1 local key = Utils.normalizeKey("w") -- "KeyW" local hit = Utils.matchModifiers(snap, { ctrl = true, shift = false }) ``` ## Notes - Pure functions — no state, no side effects. - Mouse-button indexing is 0-based and matches `Zin.state.mouseButtonDown(idx)` (0=left, 1=right, 2=middle). - `matchModifiers` ignores fields set to `nil` in `mods` — only `true`/`false` are constraints. - `normalizeKey` rejects any other single character: no bound key code is one character long, so `"?"` raises rather than resolving to a phantom key nothing consumes.

# virtual Per-tick virtual-control values. On-screen controls (and tests) write stick vectors, button states, and drag deltas here; the `touchStick` / `touchButton` / `touchDrag` / `touchPinch` binding kinds in `Zin.bindings` read them. Backs `Zin.virtual`. Three families, each keyed by the control's binding identity: - **Sticks** — zone string (`"left"`, `"right"`, ...). A vector clamped to `[-1, 1]` per component, persisting until set again or reset. - **Buttons** — id string: a `touchButton`'s `id` (stamped with the control's name when its map activates), else its `label`, else its `zone`. A held boolean; `buttonPressed` / `buttonReleased` derive edges from the previous frame boundary. - **Drags** — zone string. A delta vector that accumulates within a tick (`addDrag` sums repeated calls) and clears at the next frame boundary — the same per-frame-delta semantics as `Zin.state.mouseDelta()`. ## Edge-at-frame-boundary semantics `_beginFrame()` shifts `_buttons` into `_buttonsPrev` and clears drag deltas. It runs once per engine frame from the `Zin.tick` coordinator's gated begin block, before any binding evals happen. On-screen overlay code (script phase) calls `setButton` earlier in the same frame it wants the press recognized; `buttonPressed` reads true starting the frame `_beginFrame` next observes the transition — matching how a real key's press edge surfaces on the frame after the OS event, not the same frame the write happened. ## Exports - `M.setStick(zone: string, x: number, y: number)` — set a stick's vector (clamped). - `M.stick(zone: string) -> (number, number)` — read a stick's `(x, y)`; `(0, 0)` when unset. - `M.setButton(id: string, held: boolean)` — set a button's held state. - `M.button(id: string) -> boolean` — held state. - `M.buttonPressed(id: string) -> boolean` — press edge this frame. - `M.buttonReleased(id: string) -> boolean` — release edge this frame. - `M.addDrag(zone: string, dx: number, dy: number)` — accumulate a drag delta. - `M.drag(zone: string) -> (number, number)` — read a zone's accumulated `(dx, dy)` this frame. - `M._beginFrame()` — internal: frame-boundary edge shift + drag clear, called by `Zin.tick`. - `M._reset()` — test-only: clear all virtual state. ## Usage ```luau local Virtual = require("@builtin::modules.zinput.virtual") Virtual.setStick("left", 0.4, -0.9) local x, y = Virtual.stick("left") Virtual.setButton("Jump", true) if Virtual.buttonPressed("Jump") then -- fires once, on the frame after the set end ```

# 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. 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()`. Globals leak across reloads. 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**.

# 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 — the bare indicators only. 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 `---`), block-scalar chomping/indentation indicators (`|-`, `|+`, `>-`, `|2`, ...), 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).

# interfaces The screen-building skill: author a UI as a styled widget tree the way you write HTML and CSS, register it, and then see it and drive it through the `gui` toolbox rather than reasoning about the tree. Covers the layout and paint model, entrance and idle animation, callbacks that fire in play mode, and the build → capture → play → iterate loop that gets a screen from blank to built.

# debugging Finding out **when** something went wrong in a running world. The job this skill covers is not "what is the current value" — the current value is downstream of the moment that matters. It is: attach an instrument to running code, keep working, and be told the instant the thing happens. It exists because the reflex it replaces does not work. Running something, waiting, and taking a capture aims at a moment you cannot see: seconds pass between the call and the frame, and adding a wait only moves the guess. Polling is the same guess with more steps. Covers: probing a function call, a source line, or a value; using a condition to name the moment rather than every step toward it; reading the state a value was taken in before trusting the value; and cancelling what you armed. The trap it spends the most words on is the expensive one — a right value read in the wrong world, where the match already ended or the scene already swapped, nothing announced it, and every number afterwards means something else.