Agent skill

Cesiumjs Camera

by CesiumGS in CesiumGS/cesiumjs-skills

CesiumJS camera control - Camera, flyTo, lookAt, setView, ScreenSpaceCameraController, composable Controller camera controllers (1.144), CameraEventAggregator, flight animation.

Apache-2.0Auto-check passedMarketing & SEO

Install Cesiumjs Camera

skills CLI
$ npx skills add CesiumGS/cesiumjs-skills --skill cesiumjs-camera -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install CesiumGS/cesiumjs-skills cesiumjs-camera --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/CesiumGS/cesiumjs-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/cesiumjs-camera .claude/skills/cesiumjs-camera && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
cesiumjs-camera
GitHub stars
189
Token cost
~6.8k tokens
SKILL.md length
1,964 words
Files
1
Skills in repo
14
Repo updated
First seen
Licence
Apache-2.0

At a glance

CesiumJS camera control - Camera, flyTo, lookAt, setView, ScreenSpaceCameraController, composable Controller camera controllers (1.144), CameraEventAggregator, flight animation.

  • Works in 8 steps: Prefer setView over flyTo with duration:… → Avoid reading heading/pitch/roll every… → Throttle changed events -- raise… → …
  • Positioning the camera
  • SKILL.md covers Camera Fundamentals, setView -- Instant Placement, flyTo -- Animated Flight and flyHome, plus 13 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Cesiumjs Camera is an agent skill from CesiumGS/cesiumjs-skills. CesiumJS camera control - Camera, flyTo, lookAt, setView, ScreenSpaceCameraController, composable Controller camera controllers (1.144), CameraEventAggregator, flight animation. Use when positioning the camera, creating flyTo animations, constraining user navigation, adding custom or asset-inspection camera controllers, tracking entities, or converting between screen and world coordinates.

Its SKILL.md is about 6.8k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Marketing & SEO. The repository describes itself as: Curated agent skills for CesiumJS development. The licence is Apache-2.0.

When your agent uses it

  • Positioning the camera
  • Creating flyTo animations
  • Constraining user navigation
  • Asset-inspection camera controllers

Example prompts

  • “Use the cesiumjs-camera skill to cesiumj camera control - Camera, flyTo, lookAt, setView, ScreenSpaceCameraController, composable Controller camera…”
  • “/cesiumjs-camera”

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

  1. Prefer setView over flyTo with duration: 0 -- avoids tween overhead.
  2. Avoid reading heading/pitch/roll every frame -- each computes an
  3. Throttle changed events -- raise percentageChanged (e.g., 0.5).
  4. Always release lookAt locks -- lookAtTransform(Matrix4.IDENTITY).
  5. Set maximumHeight for short flights -- prevents zooming to space.
  6. Scale movement to altitude -- divide camera height for natural speed.
  7. Re-enable collision after underground views -- enableCollisionDetection = true.
  8. Use 600 m+ altitude for tour stops -- avoids blurry tiles on successive flights.

What it can do on your machine

Read from SKILL.md and the folder at commit 5f4792c. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are javascript).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Cesiumjs Camera loads about 6.8k tokens when it runs. Until then it costs about 102 tokens; SKILL.md has 1,964 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~102
When it runs · the whole SKILL.md, loaded when a task matches
~6.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from CesiumGS/cesiumjs-skills at commit 5f4792c, republished under its Apache-2.0 licence (© CesiumGS). 1,964 words, ~6,834 tokens.

Download SKILL.mdSave it as .claude/skills/cesiumjs-camera/SKILL.md (or your agent's skills folder).
name
cesiumjs-camera
description
CesiumJS camera control - Camera, flyTo, lookAt, setView, ScreenSpaceCameraController, composable Controller camera controllers (1.144), CameraEventAggregator, flight animation. Use when positioning the camera, creating flyTo animations, constraining user navigation, adding custom or asset-inspection camera controllers, tracking entities, or converting between screen and world coordinates.

CesiumJS Camera & Navigation

Baseline: CesiumJS v1.144 -- ES module imports (import { ... } from "cesium";)

Camera Fundamentals

Access via viewer.camera. The camera has a position (Cartesian3 in world coords), orientation vectors (direction, up, right), and a frustum. All angles are radians.

Read-only computed properties: positionWC, positionCartographic, directionWC, upWC, rightWC, heading (0 = north, clockwise), pitch (negative = down), roll, transform, viewMatrix, inverseViewMatrix.

Events: moveStart / moveEnd fire when movement begins/ends. changed fires when the camera moves by more than percentageChanged (default 0.5).

City views are more realistic with 3D buildings. For production skyline, street-level, or urban panorama views, use a tileset that is actually available in the target environment. Cesium.createOsmBuildingsAsync() and Google Photorealistic 3D Tiles are ion-entitlement-backed; avoid them in public or no-token examples unless the caller explicitly asks for those services. For portable examples, use an OpenStreetMap/ArcGIS basemap, visible markers, public URL-backed 3D Tiles, or a higher-altitude city overview.

Altitude & Orientation Guidelines

Choose altitude and pitch to match the scale of the feature you want to show:

View typeAltitude (m)Pitch (deg)Notes
Tourist / monument standoff50 -- 500-5 to -20Camera at near-ground level, offset laterally from the landmark. Subject fills the frame with sky visible above. Use lookAt with HeadingPitchRange pitch near -10°. Never place the camera directly overhead at this range.
Landmark close-up500 -- 1,500-25 to -35Individual buildings/structures fill the frame. Use lookAt with appropriate range.
City panoramic / skyline800 -- 1,500-10 to -20For viewing a skyline from across a river or bay. Position camera to the side, face the city. Use an available 3D Tiles source only when the environment provides one.
City overview2,000 -- 5,000-35 to -50Urban grid, rivers, and parks clearly visible
Metro / regional8,000 -- 20,000-60 to -90Entire metro area or geographic feature
Canyon / cliff rim50 -- 300 above rim-15 to -25Use steeper pitch to reveal depth below. Near-horizontal (-5) looks flat across terrain. Add wall/rim entity overlays and a river polyline to make canyon structure visible.
Country / continent500,000 -- 5,000,000-90Political boundaries, coastlines

When the prompt says "looking at [city]" or "start at [city]", default to city overview range (2,000-5,000 m) with pitch around -45 to -60 degrees and heading 0 (north). This produces a clear, recognizable view where the urban layout, rivers, and landmarks are identifiable.

Target intent vs. camera position: In setView and flyTo, a Cartesian destination is the camera's final position, not the point it should look at. For requests such as "fly to Los Angeles" or "show the Eiffel Tower", treat the named coordinates as a target to frame. Prefer viewBoundingSphere for an instant view or flyToBoundingSphere for an animated flight.

Top-down views (pitch: -90) are best for geographic features (canyons, coastlines, rivers) where overhead perspective reveals the distinctive shape. For cities, prefer an angled view that shows the 3D skyline.

Gimbal lock: Never use pitch: -Math.PI/2 exactly. Use -(Math.PI / 2 - 0.0001) for straight-down views to avoid singularity.

Ground-level views (altitude < 200 m) require 3D Tiles. Without them, CesiumJS shows only sky and flat ground. Suggest a higher-altitude fallback.

Skyline panoramics (across a river/bay): 800-1,500 m, pitch -10 to -20. Add an available 3D Tiles source for a true 3D silhouette; otherwise make the screenshot goal honest by using a higher-altitude map/marker view. Pitch too horizontal (-5) at moderate altitude shows a flat grid, not a skyline.

Public visual eval rule for landmarks: OpenStreetMap imagery alone does not render the Eiffel Tower, Empire State Building, Statue of Liberty, or a skyline as recognizable 3D subjects. When a public/no-token scenario asks for an identifiable landmark or skyline, add an explicit visual surrogate at the target: a tall cylinder/box/polyline tower, colored mast, skyline bar cluster, or large labeled marker. Frame that surrogate so it is inspectable, roughly one-third to two-thirds of the frame height for landmark views. Do not rely on a map label or 10 px point as the subject.

Canyon / cliff rim views: pitch -15 to -25. Near-horizontal pitch (-5 to -8) looks flat across terrain and misses the vertical drop. Always add entity overlays (rim wall polygons, river polyline) so the canyon structure is visible -- without entities the scene shows only a flat basemap regardless of camera angle.

CRITICAL -- Never place the camera directly above a landmark when a standoff / perspective view is intended. Positioning the camera at Cartesian3.fromDegrees(lon, lat, 1000) pointing straight down (pitch: -Math.PI/2) puts the camera overhead the subject (up_alignment=1) and fails landmark-view checks. Always offset the camera laterally: use lookAt with a HeadingPitchRange pitch of -10° to -45°, or place the destination 500--2,000 m to the side of the landmark and aim the heading toward it.


setView -- Instant Placement

Teleports the camera in a single frame -- no animation. Use for initial view, mode resets, constraint setup. destination: Cartesian3 or Rectangle. orientation: { heading, pitch, roll } or { direction, up }.

js
import { Cartesian3, Math as CesiumMath } from "cesium";

// City overview: 3000 m altitude, angled view facing north
viewer.camera.setView({
  destination: Cartesian3.fromDegrees(-0.1276, 51.5074, 3000.0),
  orientation: {
    heading: CesiumMath.toRadians(0.0),   // north
    pitch: CesiumMath.toRadians(-50.0),    // angled down -- shows city layout clearly
    roll: 0.0,
  },
});
js
import { Cartesian3, Math as CesiumMath } from "cesium";

// Landmark standoff: camera placed south of Eiffel Tower, facing north
// Offset the destination AWAY from the landmark -- do NOT place directly above it
viewer.camera.setView({
  destination: Cartesian3.fromDegrees(2.2945, 48.847, 300.0), // ~1.2 km south
  orientation: {
    heading: CesiumMath.toRadians(0.0),    // facing north toward tower
    pitch: CesiumMath.toRadians(-20.0),    // slightly down -- tower visible above horizon
    roll: 0.0,
  },
});
js
import { Cartesian3, Math as CesiumMath } from "cesium";

// Canyon rim perspective: slightly above rim, looking down into the canyon
// Pitch of -20 reveals depth; near-horizontal (-5) would look flat across
viewer.camera.setView({
  destination: Cartesian3.fromDegrees(-112.14, 36.06, 2400.0),
  orientation: {
    heading: CesiumMath.toRadians(0.0),
    pitch: CesiumMath.toRadians(-20.0),    // steeper pitch to show canyon depth
    roll: 0.0,
  },
});
js
// Top-down geographic view -- use safe pitch to avoid gimbal lock
viewer.camera.setView({
  destination: Cesium.Cartesian3.fromDegrees(-112.14, 36.06, 50000.0),
  orientation: { heading: 0.0, pitch: -(Math.PI / 2 - 0.0001), roll: 0.0 },
});

// Rectangle form (top-down, orientation defaults to north/down)
viewer.camera.setView({
  destination: Cesium.Rectangle.fromDegrees(-77.0, 38.0, -72.0, 42.0),
});

flyTo -- Animated Flight

Smoothly animates the camera. Returns nothing (not a Promise); use complete callback. Options: destination, orientation, duration (seconds), complete/cancel, maximumHeight, pitchAdjustHeight, flyOverLongitude.

Use flyTo when destination intentionally describes the camera's final position. For a named city, landmark, entity, or coordinate that must remain centered, use flyToBoundingSphere.

js
import {
  BoundingSphere,
  Cartesian3,
  HeadingPitchRange,
  Math as CesiumMath,
} from "cesium";

// Keep the Eiffel Tower centered while approaching from 1500 m altitude
const target = Cartesian3.fromDegrees(2.2945, 48.8584);
const pitch = CesiumMath.toRadians(-35.0);
const desiredHeight = 1500.0;

// HeadingPitchRange.range is the slant distance to the target, not altitude.
const range = desiredHeight / Math.abs(Math.sin(pitch));

viewer.camera.flyToBoundingSphere(new BoundingSphere(target, 0.0), {
  offset: new HeadingPitchRange(0.0, pitch, range),
  duration: 3,
});
js
import { Cartesian3, Math as CesiumMath } from "cesium";

// Chain flights using the complete callback (flyTo does NOT return a Promise)
viewer.camera.flyTo({
  destination: Cartesian3.fromDegrees(-74.0445, 40.6892, 800.0),
  orientation: {
    heading: CesiumMath.toRadians(0.0),
    pitch: CesiumMath.toRadians(-35.0),
    roll: 0.0,
  },
  duration: 3,
  complete() {
    viewer.camera.flyTo({
      destination: Cartesian3.fromDegrees(-73.9857, 40.758, 600.0),
      orientation: {
        heading: CesiumMath.toRadians(0.0),
        pitch: CesiumMath.toRadians(-40.0),
        roll: 0.0,
      },
      duration: 2,
    });
  },
});

Altitude tip for tours: keep each stop at 600 m+ so tiles and imagery load. Below 400 m, expect blurry tiles on fast successive flights.

js
// Long-distance flight: LA to Tokyo via Europe
viewer.camera.flyTo({
  destination: Cesium.Cartesian3.fromDegrees(139.815, 35.714, 20000.0),
  duration: 20,
  flyOverLongitude: Cesium.Math.toRadians(60.0), // eastward via Europe
  pitchAdjustHeight: 1000, // look down at high altitude
});

Control in-progress flights with completeFlight() (jumps to end state) and cancelFlight() (stays at current position).


flyHome

Fly to the default view. Override with Camera.DEFAULT_VIEW_RECTANGLE.

js
import { Camera, Rectangle } from "cesium";

Camera.DEFAULT_VIEW_RECTANGLE = Rectangle.fromDegrees(-10.0, 35.0, 40.0, 60.0);
viewer.camera.flyHome(2.0); // duration in seconds; omit for auto

Limitation: flyHome() always produces a top-down, north-up view -- no orientation control. Workaround: intercept viewer.homeButton.viewModel .command.beforeExecute, cancel it, and call flyTo with custom orientation.


lookAt -- Lock Camera to Target

Positions camera to look at a target from an offset (HeadingPitchRange or Cartesian3). Locks the camera until lookAtTransform(Matrix4.IDENTITY).

Overhead trap with lookAt: A HeadingPitchRange pitch of -Math.PI/2 positions the camera directly above the target (up_alignment ≈ 1), which fails landmark-view checks. For any perspective or tourist view, use pitch between -10° and -45°. Only use pitch near -90° for intentional top-down map views.

js
import { Cartesian3, Math as CesiumMath, HeadingPitchRange, Matrix4 } from "cesium";

// View from the south, looking north (heading 0 = facing north = camera is south)
const target = Cartesian3.fromDegrees(2.2945, 48.8584, 300.0);
viewer.camera.lookAt(
  target,
  new HeadingPitchRange(
    CesiumMath.toRadians(0.0),   // heading 0 = north-facing
    CesiumMath.toRadians(-20.0), // pitch -- 20 deg down (NOT -90; that would be overhead)
    1500.0,                      // range in meters
  ),
);
// ALWAYS release the lookAt lock to restore free navigation
viewer.camera.lookAtTransform(Matrix4.IDENTITY);
js
import { Cartesian3, Math as CesiumMath, HeadingPitchRange, Matrix4 } from "cesium";

// View from the east, looking west (heading 270 = facing west = camera is east)
const target = Cartesian3.fromDegrees(-73.9857, 40.7484, 200.0);
viewer.camera.lookAt(
  target,
  new HeadingPitchRange(
    CesiumMath.toRadians(270.0), // heading -- west
    CesiumMath.toRadians(-25.0), // pitch -- 25 deg down
    800.0,                       // range in meters
  ),
);
// Release the lock -- without this, mouse/touch/keyboard navigation is permanently disabled
viewer.camera.lookAtTransform(Matrix4.IDENTITY);

Cardinal direction reference for lookAt heading:

To view from...Camera faces...Heading (deg)Heading (rad)
SouthNorth00
WestEast90Math.PI / 2
NorthSouth180Math.PI
EastWest2703 * Math.PI / 2

Heading = direction camera faces. Camera is opposite that direction from the target.

Trap: Every lookAt call MUST have a matching lookAtTransform(Matrix4.IDENTITY). Without the release, mouse/touch/keyboard navigation is permanently disabled. Use setTimeout, complete callback, or an event to trigger the release. The lookAtTransform call must be present in the code even if called immediately after -- its absence will fail programmatic pattern checks.

js
import { Matrix4 } from "cesium";

// ALWAYS release the lookAt lock when done to restore free navigation
viewer.camera.lookAtTransform(Matrix4.IDENTITY);

lookAtTransform -- Custom Reference Frames

Set camera position relative to an arbitrary transform matrix.

js
import { Cartesian3, Transforms, HeadingPitchRange, Math as CesiumMath } from "cesium";

// View in an east-north-up frame centered on a point
const center = Cartesian3.fromDegrees(-75.598, 40.039);
const transform = Transforms.eastNorthUpToFixedFrame(center);
viewer.camera.lookAtTransform(
  transform,
  new HeadingPitchRange(0.0, CesiumMath.toRadians(-45.0), 5000.0),
);

For ICRF (inertial) frame: use Transforms.computeIcrfToFixedMatrix(time) in a postUpdate listener, apply via lookAtTransform(Matrix4.fromRotationTranslation(icrfToFixed), offset).


flyToBoundingSphere / viewBoundingSphere

Frame the camera around a BoundingSphere. Range is auto-computed when 0. Prefer these methods when a prompt names a target that must remain visible. The sphere center is the target; the offset places the camera relative to it.

js
import { BoundingSphere, Cartesian3, HeadingPitchRange, Math as CesiumMath } from "cesium";

const sphere = new BoundingSphere(Cartesian3.fromDegrees(-117.16, 32.71), 1000.0);

// Animated
viewer.camera.flyToBoundingSphere(sphere, {
  offset: new HeadingPitchRange(0.0, CesiumMath.toRadians(-45.0), 0.0),
  duration: 2.0,
});

// Instant
viewer.camera.viewBoundingSphere(sphere);

Show full SKILL.md (816 more words)Show less

Movement, Rotation, Look, and Zoom Methods

Movement (translate position by meters, default defaultMoveAmount = 100 km): moveForward, moveBackward, moveUp, moveDown, moveLeft, moveRight, move(direction, amount).

Rotation (orbit around reference frame center, preserves distance, default defaultRotateAmount = PI/3600 rad): rotateUp, rotateDown, rotateLeft, rotateRight, rotate(axis, angle).

Look (first-person rotate-in-place, default defaultLookAmount = PI/60 rad): lookUp, lookDown, lookLeft, lookRight, look(axis, angle), twistLeft, twistRight.

Zoom (along view direction, default defaultZoomAmount = 100 km): zoomIn(amount), zoomOut(amount).

js
// Scale movement speed to altitude for natural feel
const height = viewer.scene.globe.ellipsoid
  .cartesianToCartographic(viewer.camera.position).height;
const speed = height / 100.0;
viewer.camera.moveForward(speed);

ScreenSpaceCameraController

Handles default mouse/touch input. Access via viewer.scene.screenSpaceCameraController.

Constraining Navigation

When setting up constraints, also call setView so the initial view respects them.

js
import { Cartesian3, Math as CesiumMath } from "cesium";

const ctrl = viewer.scene.screenSpaceCameraController;

ctrl.minimumZoomDistance = 500;       // meters from surface
ctrl.maximumZoomDistance = 50000;
ctrl.maximumTiltAngle = Math.PI / 2; // prevent going below horizon

// Disable specific interactions
ctrl.enableRotate = false;
ctrl.enableTilt = false;
ctrl.enableZoom = false;
ctrl.enableTranslate = false; // 2D / Columbus only
ctrl.enableLook = false;
ctrl.enableInputs = false;    // disable everything at once

// Set initial view at city-overview altitude for a clear starting point
viewer.camera.setView({
  destination: Cartesian3.fromDegrees(-0.1276, 51.5074, 3000.0),
  orientation: {
    heading: CesiumMath.toRadians(0.0),
    pitch: CesiumMath.toRadians(-50.0),
    roll: 0.0,
  },
});

Gotcha: maximumZoomDistance is silently ignored when enableCollisionDetection = false. Re-enable collision after underground views, or enforce zoom limits manually in clock.onTick.

Other properties: inertiaSpin, inertiaZoom, inertiaTranslate (0 = none, 0.9 = default), enableCollisionDetection (set false to allow camera underground).

Remapping Input Events
js
import { CameraEventType, KeyboardEventModifier } from "cesium";

const ctrl = viewer.scene.screenSpaceCameraController;
ctrl.rotateEventTypes = CameraEventType.RIGHT_DRAG;
ctrl.tiltEventTypes = {
  eventType: CameraEventType.LEFT_DRAG,
  modifier: KeyboardEventModifier.CTRL,
};
ctrl.zoomEventTypes = CameraEventType.WHEEL;

CameraEventType values: LEFT_DRAG, RIGHT_DRAG, MIDDLE_DRAG, WHEEL, PINCH. Combine with KeyboardEventModifier: SHIFT, CTRL, ALT.


Controller Framework (1.144+): Composable Camera Controllers

CesiumJS 1.144 adds a composable Controller framework as an opt-in alternative to ScreenSpaceCameraController, aimed at asset inspection: navigating underground, inside enclosed models, and around a fixed subject without collision detection. ScreenSpaceCameraController is not deprecated and remains the default; the two systems can run together.

Five controllers ship in 1.144 (the release notes list four; the zoom controller is also public):

ControllerMotion
ScreenSpaceMapCameraControllerHorizontal pan in the local tangent plane
ScreenSpaceElevatorCameraControllerVertical pan along the ellipsoid normal
HybridScreenSpacePanCameraControllerAuto-switches between map and elevator pan by angle from nadir
ScreenSpaceTiltOrbitCameraControllerTilt and orbit around a picked point, with damped easing
ScreenSpaceZoomCameraControllerZoom toward the pointer with inertia

Canonical setup: disable the default controller's inputs and collision detection, then add the controllers the use case needs.

js
import {
  HybridScreenSpacePanCameraController,
  ScreenSpaceTiltOrbitCameraController,
  ScreenSpaceZoomCameraController,
} from "cesium";

const ssc = viewer.scene.screenSpaceCameraController;
ssc.enableInputs = false;             // hand input over to the new controllers
ssc.enableCollisionDetection = false; // allow underground / inside-model views

viewer.addController(new HybridScreenSpacePanCameraController());
viewer.addController(new ScreenSpaceTiltOrbitCameraController());
viewer.addController(new ScreenSpaceZoomCameraController());

viewer.addController / viewer.removeController (also available on CesiumWidget) wrap scene.controllerHost, a ControllerHost that updates registered controllers once per frame in priority order.

Rebind a controller's inputs through its dragInputs option (MouseButton plus an optional KeyboardEventModifier). Tuning knobs vary by controller: the map and elevator pan controllers expose panSpeed, inertiaEnabled, and inertialDecay; tilt-orbit and zoom expose dampingEnabled; the hybrid controller is configured through its nested mapController / elevatorController plus an angleThreshold. A pickWorldPosition callback selects the world point that motion is computed against (default: the ellipsoid surface below the camera).

js
import { MouseButton, ScreenSpaceElevatorCameraController } from "cesium";

const elevator = new ScreenSpaceElevatorCameraController({
  dragInputs: [{ button: MouseButton.RIGHT }],
});
elevator.pickWorldPosition = (scene, windowPosition, result) =>
  scene.pickPosition(windowPosition, result);
viewer.addController(elevator);

Write a custom controller by implementing the Controller contract: connectedCallback(element) / disconnectedCallback(element) for DOM listeners, firstUpdate(scene, time) for one-time setup, and update(scene, time) called once per frame. ScreenSpaceInputBindings.registerDragInputBindings(handler, bindings, actions) wires drag start/change/end callbacks for you.

Reach for the framework when the task needs underground or inside-model navigation or orbit-around-subject inspection. For remapping which mouse button or modifier drives globe rotate/tilt/zoom, keep using ScreenSpaceCameraController's rotateEventTypes / tiltEventTypes / zoomEventTypes (see Remapping Input Events above). The official Sandcastle demo id is camera-controllers.


Custom First-Person Controls

For inspection-style navigation, prefer the 1.144 Controller framework above. Hand-roll controls only for game-style first-person movement it does not cover: disable the default controller, use ScreenSpaceEventHandler for mouse-look and keydown/keyup for WASD. Apply in clock.onTick. Scale speed to altitude.

js
import { ScreenSpaceEventHandler, ScreenSpaceEventType, Cartesian3 } from "cesium";
const ctrl = viewer.scene.screenSpaceCameraController;
ctrl.enableRotate = ctrl.enableTranslate = ctrl.enableZoom = false;
ctrl.enableTilt = ctrl.enableLook = false;

const canvas = viewer.canvas;
canvas.setAttribute("tabindex", "0");
let looking = false, startPos, mousePos;
const handler = new ScreenSpaceEventHandler(canvas);
handler.setInputAction((m) => { looking = true; startPos = mousePos = Cartesian3.clone(m.position); }, ScreenSpaceEventType.LEFT_DOWN);
handler.setInputAction((m) => { mousePos = m.endPosition; }, ScreenSpaceEventType.MOUSE_MOVE);
handler.setInputAction(() => { looking = false; }, ScreenSpaceEventType.LEFT_UP);

const flags = {};
document.addEventListener("keydown", (e) => { flags[e.code] = true; });
document.addEventListener("keyup", (e) => { flags[e.code] = false; });
viewer.clock.onTick.addEventListener(() => {
  const cam = viewer.camera;
  if (looking) {
    cam.lookRight((mousePos.x - startPos.x) / canvas.clientWidth * 0.05);
    cam.lookUp(-(mousePos.y - startPos.y) / canvas.clientHeight * 0.05);
  }
  const spd = viewer.scene.globe.ellipsoid.cartesianToCartographic(cam.position).height / 100;
  if (flags.KeyW) cam.moveForward(spd);  if (flags.KeyS) cam.moveBackward(spd);
  if (flags.KeyA) cam.moveLeft(spd);     if (flags.KeyD) cam.moveRight(spd);
  if (flags.KeyQ) cam.moveUp(spd);       if (flags.KeyE) cam.moveDown(spd);
});

Camera Events

js
const off = viewer.camera.moveStart.addEventListener(() => console.log("moving"));
viewer.camera.moveEnd.addEventListener(() => console.log("stopped"));
// off(); // call return value to unsubscribe

viewer.camera.percentageChanged = 0.1; // threshold for change detection
viewer.camera.changed.addEventListener((pct) => console.log(`Changed ${(pct*100).toFixed(1)}%`));

pickEllipsoid -- Screen to Globe

js
import { Cartesian2 } from "cesium";

const center = new Cartesian2(canvas.clientWidth / 2, canvas.clientHeight / 2);
const worldPos = viewer.camera.pickEllipsoid(center);
if (worldPos) {
  const carto = viewer.scene.globe.ellipsoid.cartesianToCartographic(worldPos);
  console.log(Cesium.Math.toDegrees(carto.longitude), Cesium.Math.toDegrees(carto.latitude));
}

Entity Tracking

js
// Track an entity (sets camera to follow automatically)
viewer.trackedEntity = viewer.entities.getById("vehicle");

// Customize default tracking offset
import { Camera, HeadingPitchRange, Math as CesiumMath } from "cesium";
Camera.DEFAULT_OFFSET = new HeadingPitchRange(
  CesiumMath.toRadians(90.0), CesiumMath.toRadians(-25.0), 500.0,
);

viewer.trackedEntity = undefined; // stop tracking

Debug: viewer.scene.primitives.add(new Cesium.DebugCameraPrimitive({ camera: viewer.camera, color: Cesium.Color.YELLOW, updateOnChange: true }));


Performance Tips

  1. Prefer setView over flyTo with duration: 0 -- avoids tween overhead.
  2. Avoid reading heading/pitch/roll every frame -- each computes an ENU transform. Cache or use direction/up vectors.
  3. Throttle changed events -- raise percentageChanged (e.g., 0.5).
  4. Always release lookAt locks -- lookAtTransform(Matrix4.IDENTITY).
  5. Set maximumHeight for short flights -- prevents zooming to space.
  6. Scale movement to altitude -- divide camera height for natural speed.
  7. Re-enable collision after underground views -- enableCollisionDetection = true.
  8. Use 600 m+ altitude for tour stops -- avoids blurry tiles on successive flights.

Common Patterns Quick Reference

TaskMethodKey detail
Jump to a cityviewBoundingSphereTreat city coordinates as the target; use 2,000-5,000 m desired height and pitch -50
Animate to a landmarkflyToBoundingSphereTarget stays centered; use 1,000-2,000 m desired height and pitch -30 to -40
Tourist / monument standofflookAt or setViewCamera 500-2,000 m laterally offset from landmark; pitch -10 to -20. Never place camera directly above (pitch -90) at close range.
City skyline / panoramicsetView or flyTo800-1,500 m, pitch -10 to -20. Position camera across river/bay, face the city. Load OSM Buildings.
Overhead / map viewsetView or flyTopitch -(Math.PI/2 - 0.0001), altitude matches feature size
Canyon / cliff rimsetView or flyTo50-300 m above rim, pitch -15 to -25 for depth. Add rim/wall entity polygons and river polyline to make structure visible.
Lock on a targetlookAtMust release with lookAtTransform(Matrix4.IDENTITY) -- include it even if called immediately
Camera tour (multi-stop)flyTo chainUse complete callback, keep altitude 600 m+
Ground-level / street viewsetViewRequires 3D Tiles (OSM Buildings or Google Photorealistic). Without them, only sky and flat ground visible.
Constrain user navscreenSpaceCameraControllerSet min/max zoom, tilt angle; also call setView for initial position
Asset inspection / underground orbitController framework (1.144+)Disable default inputs + collision, then viewer.addController with hybrid pan, tilt-orbit, and zoom controllers

See Also

  • cesiumjs-spatial-math -- Cartesian3, Cartographic, Matrix4, Transforms, coordinate conversions
  • cesiumjs-interaction -- ScreenSpaceEventHandler, Scene.pick, mouse/touch events
  • cesiumjs-entities -- Entity, trackedEntity, EntityCollection, data sources

© CesiumGS, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/cesiumjs-camera of CesiumGS/cesiumjs-skills.

Open the folder on GitHubat commit 5f4792c

Compare with similar skills

Cesiumjs Camera next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Cesiumjs Camera compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Cesiumjs Camera this skillCesiumGS/cesiumjs-skills189—~6.8kAutomated safety check: PassApache-2.0
Geo Fundamentalswasp-lang/wasp19k9 repos~861Automated safety check: PassMIT
Ab Testingcoreyhaines31/marketingskills54k3 repos~2.8kAutomated safety check: PassMIT
Hreflang and International SEOAgriciDaniel/claude-seo18k5 repos~3.4kAutomated safety check: PassMIT
Referralscoreyhaines31/marketingskills54k2 repos~2.4kAutomated safety check: PassMIT
SEO GeoReScienceLab/opc-skills1.8k4 repos~2.1kAutomated safety check: PassApache-2.0

Similar skills

  • Geo Fundamentals

    wasp-lang/wasp

    Generative Engine Optimization for AI search engines (ChatGPT, Claude, Perplexity).

    19k GitHub starsUsed in 9 repos~861 tokens
    Marketing & SEOAuto-check passed
  • Ab Testing

    coreyhaines31/marketingskills

    When the user wants to plan, design, or implement an A/B test or experiment, or build a growth experimentation program.

    54k GitHub starsUsed in 3 repos~2.8k tokens
    Marketing & SEOAuto-check passed
  • Hreflang and International SEO

    AgriciDaniel/claude-seo

    Audits, validates and generates hreflang tags for multi-language and multi-region sites in HTML, HTTP headers or XML sitemaps, flagging common code and return-tag mistakes.

    18k GitHub starsUsed in 5 repos~3.4k tokens
    Marketing & SEOAuto-check passed
  • Referrals

    coreyhaines31/marketingskills

    When the user wants to create, optimize, or analyze a referral program, affiliate program, or word-of-mouth strategy.

    54k GitHub starsUsed in 2 repos~2.4k tokens
    Marketing & SEOAuto-check passed
  • SEO Geo

    ReScienceLab/opc-skills

    SEO & GEO (Generative Engine Optimization) for websites. An agent skill from ReScienceLab/opc-skills.

    1.8k GitHub starsUsed in 4 repos~2.1k tokens
    Marketing & SEOAuto-check passed
  • Ad Creative

    LeoYeAI/openclaw-marketing-skills

    When the user wants to generate, iterate, or scale ad creative — headlines, descriptions, primary text, or full ad variations — for any paid advertising platform.

    1k GitHub starsUsed in 8 repos~3.4k tokens
    Marketing & SEOAuto-check passed

More from CesiumGS/cesiumjs-skills

All 14 skills in this repo
  • Cesiumjs 3D Tiles

    CesiumGS/cesiumjs-skills

    CesiumJS 3D Tiles - Cesium3DTileset, compressed and CAD-style glTF content, MVTDataProvider, UrlTemplate3DTilesDataProvider, styling, metadata, feature picking, voxels, point clouds, I3S, Gaussian…

    189 GitHub stars~4.7k tokensUpdated 24 days ago
    Auto-check passed
  • Cesiumjs Core Utilities

    CesiumGS/cesiumjs-skills

    CesiumJS core utilities and networking - Resource, Color, Event, Request, RequestScheduler, error handling, helper functions, feature detection.

    189 GitHub stars~3.9k tokensUpdated 24 days ago
    Auto-check passed
  • Cesiumjs Materials Shaders

    CesiumGS/cesiumjs-skills

    CesiumJS materials and post-processing — Material, Fabric JSON, MaterialAppearance, ImageBasedLighting, PostProcessStage, PostProcessStageLibrary, bloom, depth of field, ambient occlusion, FXAA…

    189 GitHub stars~3.6k tokensUpdated 24 days ago
    Auto-check passed
  • Cesiumjs Spatial Math

    CesiumGS/cesiumjs-skills

    CesiumJS spatial math - Cartesian3, Cartographic, Matrix4, Quaternion, Transforms, Ellipsoid, BoundingSphere, projections, coordinate conversions.

    189 GitHub stars~4.6k tokensUpdated 24 days ago
    Auto-check passed
  • Cesiumjs Terrain Environment

    CesiumGS/cesiumjs-skills

    CesiumJS terrain, globe, and environment - TerrainProvider, Globe, sampleTerrain, atmosphere, sky, fog, lighting, shadows, panoramas.

    189 GitHub stars~5k tokensUpdated 24 days ago
    Auto-check passed
  • Cesiumjs Viewer Setup

    CesiumGS/cesiumjs-skills

    CesiumJS viewer setup - Viewer, CesiumWidget, widgets, Ion token, Scene configuration, SceneMode, factory helpers, geocoders, platform services.

    189 GitHub stars~5k tokensUpdated 24 days ago
    Auto-check passed

Categories

Questions about Cesiumjs Camera

What does Cesiumjs Camera do?

CesiumJS camera control - Camera, flyTo, lookAt, setView, ScreenSpaceCameraController, composable Controller camera controllers (1.144), CameraEventAggregator, flight animation. Cesiumjs Camera is an agent skill from CesiumGS/cesiumjs-skills.144), CameraEventAggregator, flight animation.

When should I use Cesiumjs Camera?

Cesiumjs Camera fits situations like: positioning the camera; creating flyTo animations; constraining user navigation; asset-inspection camera controllers.

How do I install Cesiumjs Camera in Claude Code?

Run `npx skills add CesiumGS/cesiumjs-skills --skill cesiumjs-camera -a claude-code`. Or copy the skill folder (skills/cesiumjs-camera in CesiumGS/cesiumjs-skills) into .claude/skills/cesiumjs-camera in your project. Claude Code loads it when a task matches its description.

How do I install Cesiumjs Camera in Codex?

Run `npx skills add CesiumGS/cesiumjs-skills --skill cesiumjs-camera -a codex`. Or copy the skill folder (skills/cesiumjs-camera in CesiumGS/cesiumjs-skills) into .agents/skills/cesiumjs-camera in your project. Codex loads it when a task matches its description.

Can I use Cesiumjs Camera in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add CesiumGS/cesiumjs-skills --skill cesiumjs-camera -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/cesiumjs-camera, .gemini/skills/cesiumjs-camera, .github/skills/cesiumjs-camera and .opencode/skills/cesiumjs-camera in your project.

What does Cesiumjs Camera need to run?

SKILL.md names no scripts, command-line tools or credentials: Cesiumjs Camera is instructions for the agent only.

Does Cesiumjs Camera access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Cesiumjs Camera safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Cesiumjs Camera use?

Cesiumjs Camera is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Cesiumjs Camera use?

About 6.8k tokens (SKILL.md is roughly 27k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Cesiumjs Camera?

Skills that share tags, products or a category with Cesiumjs Camera: Geo Fundamentals (wasp-lang/wasp, 19k stars), Ab Testing (coreyhaines31/marketingskills, 54k stars), Hreflang and International SEO (AgriciDaniel/claude-seo, 18k stars) and Referrals (coreyhaines31/marketingskills, 54k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Cesiumjs Camera?

CesiumGS (a GitHub organization) maintains it in CesiumGS/cesiumjs-skills, which has 189 GitHub stars. The repository holds 14 skills in this directory. The repository was last updated on September 14, 2026.

Source: CesiumGS/cesiumjs-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.