Generated banner for Display Glasses With Jetpack Compose Glimmer
Generated logo for Display Glasses With Jetpack Compose Glimmer

Display Glasses With Jetpack Compose Glimmer

Android (Google)OfficialCreated by Google LLCDistributed by Android (Google)

  • Skill
  • Tool integration
  • Software development
  • Apache-2.0
  • No issues found

From the author

Provides guidelines for developing projected Android XR apps for display glasses using the Jetpack Compose Glimmer UI toolkit. Covers Glimmer design principles, workflows, and interaction models for the glasses form factor. Use this skill to build an Android XR Augmented Experience app with Jetpack Compose Glimmer. Don't use for audio-only AI glasses without a display, non-glasses Android XR headsets (spatial panels or 3D volumes with Jetpack Compose for XR), or phone and tablet Material 3 UIs.
— the description by Google LLC

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

This skill provides design principles, UI patterns, and development workflows for creating projected Android XR applications for display glasses. It focuses on the Jetpack Compose Glimmer toolkit, covering component usage, theming, input handling, and hardware integration for additive display environments.

What it does

  • Defines Glimmer design principles and interaction models for display glasses.
  • Provides guidance on setting up projected activities and managing device connectivity.
  • Outlines UI component usage including cards, buttons, lists, and stacks.
  • Details typography, color, and depth token usage for additive displays.
  • Explains mapping hardware touchpad and gesture inputs to app actions.
  • Offers patterns for voice input indicators and hardware access.

When to use it

  • When developing Android XR apps specifically for display glasses.
  • When implementing UI using the Jetpack Compose Glimmer toolkit.
  • When designing interactions for one-dimensional hardware controls.
  • When configuring apps to run as projected activities on connected eyewear.

Advantages

  • Ensures consistent UI behavior across display glasses.
  • Provides specific guidance for additive display rendering.
  • Standardizes interaction patterns for limited input hardware.
  • Facilitates proper integration with the Android XR projected context.

Usage examples

“How do I set up a basic projected activity for display glasses?”
The skill provides the AndroidManifest configuration and the necessary Activity setup, including the use of ProjectedDisplayController and GlimmerTheme.
“What are the best practices for card layouts in Glimmer?”
It explains how to use ActionCard for single actions and ButtonGroup with non-focusable cards to avoid focus contention.
“How should I handle user input on display glasses?”
The skill details mapping touchpad gestures like taps and swipes, and explains how to use System Back for navigation.
“How do I implement a voice input indicator?”
It provides a code pattern for using the VoiceInputIndicator component driven by SpeechRecognizer RMS changes.

Install

  1. Download the skill as a ZIP:

    https://mp-staging.autogineer.com/agent-skills/android/display-glasses-with-jetpack-compose-glimmer/download
  2. Unzip it into the cross-client skills folder of your project:

    mkdir -p .agents/skills && unzip display-glasses-with-jetpack-compose-glimmer.zip -d .agents/skills

Documentation by Google LLC

Glossary

TermDefinition
Intelligent eyewearHands-free eyewear with speakers, camera, and microphone (audio-only or display glasses).
Display glassesEyewear with a private in-lens display for glanceable visuals that pair with audio.
Jetpack Compose GlimmerCompose UI toolkit providing components, theming, and focus behavior for display glasses.
Projected activityAn Activity running on a phone host whose UI and input are projected to connected eyewear.
Projected deviceAn XR eyewear device connected to an Android host phone that projects app content to it.
GlimmerThemeRoot styling provider for GlimmerColors, GlimmerTypography, and GlimmerShapes.
Additive displayDisplay where black (#000000) renders as 100% transparent by adding light to the scene.
Augmented experiencesLightweight, additive Android XR experiences that help users while they're on the go.
Visual anglePerceived size in XR; minimum readable text is 0.6 degrees (about 18sp at 1 m).

Prerequisites

  • Target compileSdk 37 or higher (compileSdk = 37 or compileSdk { version = release(37) }). Update compileSdk to 37 if lower.

  • When adding dependencies, follow the instructions in Create your first activity for intelligent eyewear:

    • androidx.xr.glimmer:glimmer (minimum 1.0.0-alpha21, or catalog alias libs.androidx.glimmer)

    • androidx.xr.glimmer:glimmer-google-fonts (minimum 1.0.0-alpha21, or catalog alias libs.androidx.glimmer.google.fonts)

    • androidx.xr.projected:projected (minimum 1.0.0-alpha13, or catalog alias libs.androidx.xr.projected or libs.androidx.projected)

    • Check gradle/libs.versions.toml first; if catalog entries exist, use them in app/build.gradle.kts and verify their versions meet the minimums. Never guess 1.0.0-alpha01 for glimmer-google-fonts.

Core constraints and limitations

  • Don't : Use MaterialTheme or Material 3 components (androidx.compose.material3.*) inside Projected Activities or Glimmer UI.

  • Do : Use GlimmerTheme and Glimmer components (androidx.xr.glimmer.*).

  • Do : Always pass createGoogleSansFlexTypography() from androidx.xr.glimmer.googlefonts as typography in GlimmerTheme.

  • Specific hardware device models or unique capabilities aren't detailed.

  • Elevation : Material 3 shadow or elevation modifiers aren't supported. Use ShadowScope or Depth tokens from Jetpack Compose Glimmer.

  • Typography limits: Minimum readable text size is 18sp (0.6°). Avoid "Thin" or "Hairline" weights, which shimmer on additive lenses.

1. Set up dependencies

  • Set up projected activity : When creating a projected activity, follow Create your first activity for intelligent eyewear and projectedcontext-source.md. Set android:requiredDisplayCategory to android.hardware.display.category.XR_PROJECTED in AndroidManifest.xml:

    <activity
        android:name="com.example.xr.projected.ProjectedMainActivity"
        android:exported="true"
        android:requiredDisplayCategory="android.hardware.display.category.XR_PROJECTED">
        <intent-filter>
            <action android:name="android.intent.action.MAIN" />
            <category android:name="android.intent.category.XR_PROJECTED_LAUNCHER"/>
        </intent-filter>
    </activity>

    <br />

  • Mobile app integration : In the host phone app, add a button with Text("Launch on Glasses"). Observe ProjectedContext.isProjectedDeviceConnected(context, coroutineContext) using collectAsStateWithLifecycle(initialValue = false) to set the button's enabled state so it's disabled when glasses are disconnected. When clicked, launch GlassesMainActivity on the glasses display using ProjectedContext.createProjectedActivityOptions(context).toBundle() as shown in the following snippet:

    val options = ProjectedContext.createProjectedActivityOptions(context)
    val intent = Intent(context, ProjectedMainActivity::class.java)
    context.startActivity(intent, options.toBundle())

    <br />

  • UI library and font : When declaring the Glimmer, Google Fonts, and Projected artifacts in app/build.gradle.kts, follow Declaring Jetpack Compose Glimmer Dependencies, and follow glimmersansflextypography-source.md when configuring Google Sans Flex using createGoogleSansFlexTypography.

  • Theming, initial focus, and black background : On additive displays, pure black (Color.Black or GlimmerTheme.colors.background) renders as 100% transparent. In onCreate (annotated with @OptIn(ExperimentalComposeUiApi::class)), set ComposeUiFlags.isInitialFocusOnFocusableAvailable = true (androidx.compose.ui.ComposeUiFlags), wrap content in GlimmerTheme(typography = createGoogleSansFlexTypography()), and apply Modifier.background(Color.Black) (or GlimmerTheme.colors.background) to the root container:

    @OptIn(ExperimentalProjectedApi::class)
    class ProjectedMainActivity : ComponentActivity() {
    
        private var displayController: ProjectedDisplayController? = null
        private var isVisualUiSupported by mutableStateOf(false)
        private var areVisualsOn by mutableStateOf(true)
    
        override fun onCreate(savedInstanceState: Bundle?) {
            super.onCreate(savedInstanceState)
    
            lifecycle.addObserver(object : DefaultLifecycleObserver {
                override fun onDestroy(owner: LifecycleOwner) {
                    displayController?.close()
                    displayController = null
                }
            })
    
            initializeGlassesFeatures()
    
            setContent {
                GlimmerTheme {
                    HomeScreen(
                        areVisualsOn = areVisualsOn,
                        isVisualUiSupported = isVisualUiSupported,
                        onClose = { finish() }
                    )
                }
            }
        }
    
        private fun initializeGlassesFeatures() {
            lifecycleScope.launch {
                // Check device capabilities
                val projectedDeviceController = ProjectedDeviceController.create(this@ProjectedMainActivity)
                isVisualUiSupported = projectedDeviceController.capabilities.contains(CAPABILITY_VISUAL_UI)
    
                val controller = ProjectedDisplayController.create(this@ProjectedMainActivity)
                displayController = controller
                val observer = GlassesLifecycleObserver(
                    context = this@ProjectedMainActivity,
                    controller = controller,
                    onVisualsChanged = { visualsOn -> areVisualsOn = visualsOn }
                )
                lifecycle.addObserver(observer)
            }
        }
    }

    <br />

  • Hardware capabilities, permissions, and access : When checking hardware capabilities, requesting permissions, or accessing glasses hardware, follow Check device capabilities at runtime, Request hardware permissions, and Use a projected context to access hardware.

2. Minimize and translate the UI

  • Build UI with Jetpack Compose Glimmer components and use depth for priority.

  • Design from the bottom up and always bottom-align UI (Alignment.BottomCenter).

  • One thing at a time : Show one primary piece of information at a time (for example, using VerticalStack) and avoid multiple simultaneous cards.

  • Color contrast : When choosing colors, follow material-hct-source.md to maintain at least a 70% tone difference between foreground and background in HCT.

3. Map input controls

  • Map tap and swipe interactions to the glasses touchpad. Inputs are one-dimensional; avoid nesting scrolling controls.

  • Use System Back to dismiss temporary states or detailed views.

  • When implementing focus navigation and hardware input handling, follow Focus in Jetpack Compose Glimmer and Hardware Controls for display glasses. Focus movement is continuous for scrollable containers and discrete for elements like button rows.

4. Build with Jetpack Compose Glimmer

Implementing Glimmer styles

When accessing theme tokens using GlimmerTheme, follow glimmertheme-source.md.

CategoryTokenValue / Role
Colorprimary / secondary#9BBFFF / #4C88E9 (Focal colors)
Colorpositive / negative#7AD48F (Confirmation) / #F57082 (Alert)
Colorbackground#000000 (Pure black, renders 100% transparent)
Colorsurface#262626 (Use SurfaceDefaults.border for borders)
ShapeStandard / SmallRoundedCornerShape(36.dp) / RoundedCornerShape(12.dp)
Typography scale (Google Sans Flex) and depth levels

Use Google Sans Flex with variable axes ROND = 100f, wdth = 100f, opsz = 9f, and wght by role:

StyleSize / Line-HeightWeight
Title Large / Medium / Small30/36.sp, 24/32.sp, 20/28.sp725
Body Large / Medium / Small30/36.sp, 24/32.sp, 20/28.sp520
Caption18.sp / 28.sp650

When simulating depth with DepthEffect and DepthEffectLevels, follow deptheffect-source.md and deptheffectlevels-source.md: use level1 for resting cards and background UI, and level2 for focused or pressed interactive UI.

Implementing Jetpack Compose Glimmer components
Cards, ActionCards, ImageCards, and ButtonGroups

Cards (androidx.xr.glimmer.Card) group related content into a single focal point. Never embed a Card within a ListItem. When implementing cards, follow the instructions in Jetpack Compose Glimmer: Cards, card-source.md, and card-samples-source.md.

  • Clickable versus non-clickable card : Pass onClick = { ... } if the entire Card acts as a trigger; omit onClick if informational. Control focus on non-clickable cards with focusable: Boolean = true (such as Card(focusable = false)).

  • Single primary action (ActionCard) : When a card has a single primary button action anchored to its bottom edge, use androidx.xr.glimmer.ActionCard instead of placing a Button inside Card. Its action button takes focus directly:

    ActionCard(
        action = {
            Button(onClick = {}, trailingIcon = { Icon(FavoriteIcon, "Localized description") }) {
                Text("Send")
            }
        }
    ) {
        Text("This is a card with a title, leading icon, header image, and action")
    }

    <br />

  • Multiple actions (Card and ButtonGroup)---avoiding focus contention : When pairing a card with multiple action buttons, use androidx.xr.glimmer.ButtonGroup (with optional rememberButtonGroupState). Never nest ButtonGroup inside Card { ... }. Place a non-focusable Card(focusable = false) (without onClick) above ButtonGroup as a sibling in a Column so the card and buttons don't compete for focus:

    ButtonGroup(modifier = Modifier.fillMaxWidth()) {
        Button(onClick = {}) { Text("Button 1") }
        Button(onClick = {}) { Text("Button 2") }
        Button(onClick = {}) { Text("Button 3") }
        Button(onClick = {}) { Text("Button 4") }
        Button(onClick = {}) { Text("Button 5") }
    }

    <br />

  • Image cards : Use ImageCard, LeadingImageCard, or TrailingImageCard to display prominent imagery alongside card text.

Buttons and ToggleButtons

When implementing buttons or toggle buttons, follow the instructions in Jetpack Compose Glimmer: Buttons, button-source.md, and button-samples-source.md.

  • Standard and icon buttons : Use Button for text actions (with optional leadingIcon/trailingIcon) and IconButton for icon-only triggers (such as media next-track or previous-track buttons).

  • Toggle buttons (ToggleButton and IconToggleButton) : Use ToggleButton or IconToggleButton for stateful on/off or play/pause controls that morph based on checked:

    @Composable
    fun ToggleButtonWithLeadingIconSample() {
        var checked by remember { mutableStateOf(false) }
    
        ToggleButton(
            checked = checked,
            leadingIcon = {
                Icon(
                    if (checked) FavoriteIcon else OutlinedFavoriteIcon,
                    contentDescription = "Toggle favorite"
                )
            },
            onCheckedChange = { checked = it }
        ) {
            Text(if (checked) "On" else "Off")
        }
    }

    <br />

Title chips

TitleChip (androidx.xr.glimmer.TitleChip) is a non-interactive pill header (three words or fewer, single line). When implementing title chips, follow the instructions in Jetpack Compose Glimmer: Title Chips, titlechip-source.md, and titlechip-samples-source.md.

  • Spatial spacing : When placing a standalone TitleChip above a Card or ActionCard, center the TitleChip horizontally and separate it from the card using TitleChipDefaults.associatedContentSpacing (using Spacer or Arrangement.spacedBy(TitleChipDefaults.associatedContentSpacing)):

    @Composable
    fun TitleChipWithCardSample() {
        Column(horizontalAlignment = Alignment.CenterHorizontally) {
            TitleChip { Text("Title Chip") }
    
            Spacer(Modifier.height(TitleChipDefaults.associatedContentSpacing))
    
            Card(
                title = { Text("Title") },
                subtitle = { Text("Subtitle") },
                leadingIcon = { Icon(FavoriteIcon, "Localized description") },
            ) {
                Text("Card Content")
            }
        }
    }

    <br />

Icons

When adding icons or icon buttons, follow the instructions in Jetpack Compose Glimmer: Icons, icon-source.md, iconbutton-source.md, and iconsizes-source.md.

  • Sizing and interactivity : Size icons with GlimmerTheme.iconSizes (small, medium, large). Use IconButton or IconToggleButton for triggers and Icon for status indicators.

  • Contrast and library : Never use pure black (Color.Black) for icon tints; black renders as transparent. Use Material Symbols (600 weight, Rounded, Unfilled). If androidx.compose.material:material-icons-* isn't in app/build.gradle.kts, build vectors with ImageVector.Builder or painterResource instead of importing androidx.compose.material.icons.*.

Lists

Use GlimmerLazyColumn (androidx.xr.glimmer.list.GlimmerLazyColumn) for homogeneous scrollable items. When building scrollable lists and list items, follow the API definitions in list-source.md, listitem-source.md, and liststate-source.md.

  • Always use GlimmerLazyColumn : Never use foundation LazyColumn. Import GlimmerLazyColumn, rememberGlimmerLazyListState, and items from androidx.xr.glimmer.list (item is a scope member, not imported). For programmatic scrolling to the first item, hoist val listState = rememberGlimmerLazyListState() and call listState.scrollToItem(0) or listState.animateScrollToItem(0).

  • Integrated title, spacing, and rows : Pass headers using title = { TitleChip { Text("...") } } directly on GlimmerLazyColumn (not outside it), set verticalArrangement = Arrangement.spacedBy(20.dp), and render rows with androidx.xr.glimmer.ListItem(onClick = { ... }, supportingLabel = { Text(...) }) { Text(...) } (pass the primary label as the trailing content lambda, not headlineContent, and never nest a Card inside ListItem):

    @Composable
    fun GlimmerLazyColumnWithTitleChipSample() {
        val ingredientItems =
            listOf("Milk", "Flour", "Egg", "Salt", "Apples", "Butter", "Vanilla", "Sugar", "Cinnamon")
        GlimmerLazyColumn(title = { TitleChip { Text("Ingredients") } }) {
            items(ingredientItems) { text -> ListItem { Text(text) } }
        }
    }

    <br />

Stacks

Use VerticalStack (androidx.xr.glimmer.stack.VerticalStack) to display heterogeneous cards one at a time (such as calendar, transit, and weather cards). Never use a TitleChip with VerticalStack. When implementing a VerticalStack, follow the API definitions in stack-source.md, stackstate-source.md, and stackitemscope-source.md.

  • Bottom alignment, height (+66.dp), and itemDecoration : Align VerticalStack to Alignment.BottomCenter, set its height to 66.dp taller than the tallest card (maxCardHeight + 66.dp), and apply Modifier.itemDecoration(CardDefaults.shape) to the Card inside every item { ... } block so background cards are clipped and masked:

    @Composable
    fun VerticalStackSample() {
        VerticalStack {
            item {
                Card { Text("Top Card") }
            }
            item {
                Card { Text("Second Card") }
            }
            item {
                Card { Text("Third Card") }
            }
        }
    }

    <br />

Horizontal pagers and animated text

Use GlimmerHorizontalPager (androidx.xr.glimmer.pager.GlimmerHorizontalPager) with rememberGlimmerPagerState(pageCount = { ... }) for horizontal card walkthroughs (never foundation HorizontalPager).

  • Smooth text motion : Inside pager pages, configure Glimmer Text (androidx.xr.glimmer.Text, never androidx.compose.material3.Text) with textMotion = TextMotion.Animated (using .copy(textMotion = TextMotion.Animated)) for smooth rendering during snap transitions:

    // Hoist the pager state, specifying the total page count with a lambda.
    val pagerState = rememberGlimmerPagerState(pageCount = { 10 })
    
    GlimmerHorizontalPager(
        state = pagerState,
        modifier = Modifier.fillMaxSize(),
    ) { page ->
        // Use Glimmer components like Card and Text for optimized glasses styling.
        Card(modifier = Modifier.fillMaxWidth()) {
            Text(
                text = "Page: $page",
                // Recommended: use TextMotion.Animated for smooth transitions in a pager.
                style = LocalTextStyle.current.copy(textMotion = TextMotion.Animated),
            )
        }
    }

    <br />

Surface

Modifier.surface (androidx.xr.glimmer.surface) builds custom interactive components (such as status badges without Card or Button). When creating custom surfaces, follow the API definitions in surface-source.md.

  • Shared MutableInteractionSource and order : Create a single remember { MutableInteractionSource() } and pass it to both Modifier.surface(interactionSource = interactionSource, ...) and .clickable(interactionSource = interactionSource, ...) (or .focusable), chaining .surface(...) before .clickable(...):

    @Composable
    fun FocusableSurfaceSample() {
        val interactionSource = remember { MutableInteractionSource() }
        Box(
            modifier = Modifier
                .size(100.dp)
                .surface(interactionSource = interactionSource)
                .focusable(interactionSource = interactionSource),
            contentAlignment = Alignment.Center
        ) {
            Text("Focusable")
        }
    }
    
    @Composable
    fun ClickableSurfaceSample() {
        val interactionSource = remember { MutableInteractionSource() }
        Box(
            modifier = Modifier
                .size(100.dp)
                .surface(interactionSource = interactionSource)
                .focusable(interactionSource = interactionSource)
                .clickable(
                    interactionSource = interactionSource,
                    indication = null
                ) { /* Handle click action */ },
            contentAlignment = Alignment.Center
        ) {
            Text("Clickable")
        }
    }

    <br />

Voice input indicator and indirect pointer gestures
  • VoiceInputIndicator and ContainedVoiceInputIndicator : Pass a level lambda returning a Float normalized to 0.0f..1.0f. When driven by SpeechRecognizer (RecognitionListener.onRmsChanged(rmsdB: Float)), normalize and clamp rmsdB with .coerceIn(0f, 1f):

    // Example state variable to hold the normalized level (0.0 to 1.0)
    // that the VoiceInputIndicator component expects.
    val audioLevel = MutableStateFlow(0f)
    
    // Initialize the Android SpeechRecognizer
    val speechRecognizer = SpeechRecognizer.createSpeechRecognizer(context)
    
    // Listener to capture speech events and audio level changes
    val listener = object : RecognitionListener {
        override fun onRmsChanged(rmsdB: Float) {
            // Normalize raw dB level to a 0.0-1.0 range.
            // Android SpeechRecognizer's rmsdB typically ranges from 0 to ~10.
            audioLevel.value = ((rmsdB - 1f) / 9f).coerceIn(0f, 1f)
        }
    
        // ... Implement other required RecognitionListener methods ...
        override fun onReadyForSpeech(params: Bundle?) {}
        override fun onBeginningOfSpeech() {}
        override fun onEndOfSpeech() {}
        override fun onError(error: Int) {}
        override fun onResults(results: Bundle?) {}
        override fun onPartialResults(partialResults: Bundle?) {}
        override fun onEvent(eventType: Int, params: Bundle?) {}
        override fun onBufferReceived(buffer: ByteArray?) {}
    }
    
    // Attach the listener to the recognizer
    speechRecognizer.setRecognitionListener(listener)
    
    // Create an intent to specify the recognition model and behavior
    val intent = Intent(RecognizerIntent.ACTION_RECOGNIZE_SPEECH).apply {
        putExtra(RecognizerIntent.EXTRA_LANGUAGE_MODEL, RecognizerIntent.LANGUAGE_MODEL_FREE_FORM)
    }
    // Begin listening for audio input
    speechRecognizer.startListening(intent)

    <br />

    @Composable
    fun VoiceInputExample(modifier: Modifier) {
        // Collect the flow as Compose State so the UI reacts to changes in real-time.
        val currentLevel by audioLevel.collectAsState()
    
        VoiceInputIndicator(
            // The VoiceInputIndicator component provides a visual "pulse" or indicator
            // that changes based on the 'level' lambda, which returns a Float between 0.0 and 1.0.
            level = { currentLevel },
            modifier = modifier
        )
    }

    <br />

  • Modifier.onIndirectPointerGesture : Handle low-level touchpad tap (onClick), forward swipe (onSwipeForward), and backward swipe (onSwipeBackward) gestures using androidx.xr.glimmer.onIndirectPointerGesture on a container made focusable (using focusTarget, focusable, or surface(focusable = true)):

    @Composable
    fun OnIndirectPointerGestureSample() {
        var lastGesture by remember { mutableStateOf("None") }
    
        Box(
            modifier =
                Modifier.fillMaxSize()
                    .onIndirectPointerGesture(
                        enabled = true,
                        onSwipeForward = { lastGesture = "Forward" },
                        onSwipeBackward = { lastGesture = "Backward" },
                        onClick = { lastGesture = "Click" },
                    )
                    .focusTarget(),
            contentAlignment = Alignment.Center,
        ) {
            Text("Last Gesture: $lastGesture")
        }
    }

    <br />

5. Integrate with system UI

Files

references/stack-source.md

When creating a Glimmer Stack component, refer to the following source code in Stack.kt:

/*
 * Copyright 2025 The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package androidx.xr.glimmer.stack

import androidx.compose.animation.core.VisibilityThreshold
import androidx.compose.animation.core.spring
import androidx.compose.foundation.layout.PaddingValues
import androidx.compose.foundation.pager.PagerDefaults
import androidx.compose.foundation.pager.PagerSnapDistance
import androidx.compose.foundation.pager.PagerState
import androidx.compose.foundation.pager.VerticalPager
import androidx.compose.runtime.Composable
import androidx.compose.runtime.derivedStateOf
import androidx.compose.runtime.referentialEqualityPolicy
import androidx.compose.runtime.remember
import androidx.compose.runtime.rememberUpdatedState
import androidx.compose.ui.Modifier
import androidx.compose.ui.draw.drawWithContent
import androidx.compose.ui.focus.FocusRequester
import androidx.compose.ui.focus.focusRequester
import androidx.compose.ui.focus.onFocusChanged
import androidx.compose.ui.geometry.Offset
import androidx.compose.ui.geometry.Size
import androidx.compose.ui.graphics.BlendMode
import androidx.compose.ui.graphics.Color
import androidx.compose.ui.input.nestedscroll.nestedScroll
import androidx.compose.ui.layout.Layout
import androidx.compose.ui.layout.onPlaced
import androidx.compose.ui.unit.dp
import androidx.compose.ui.util.fastFirstOrNull
import androidx.compose.ui.util.fastForEach
import androidx.compose.ui.util.fastMap
import androidx.compose.ui.util.lerp
import androidx.compose.ui.zIndex
import androidx.xr.glimmer.internal.SingleItemScrollConstraintConnection
import kotlin.math.roundToInt

/**
 * [VerticalStack] is a lazy, vertically scrollable layout that arranges its items in a visually
 * overlapping, three-dimensional sequence, which resembles a deck of cards. The primary item is
 * prominently displayed in the foreground. Subsequent items are positioned behind the primary item
 * along the z-axis with a small portion of the next item revealed to indicate depth and upcoming
 * content.
 *
 * As the user scrolls vertically, the active foreground item transitions out of view, allowing the
 * item immediately beneath it to slide into the prominent foreground position. Items always
 * snap-animate into the foreground position after the user's gesture ends.
 *
 * Note: When displaying text within a [VerticalStack], it is strongly recommended to set
 * [androidx.compose.ui.text.TextStyle.textMotion] to
 * [androidx.compose.ui.text.style.TextMotion.Animated]. This ensures smooth rendering during layout
 * animations or scaling transitions, preventing pixel-snapping artifacts.
 *
 * @sample androidx.xr.glimmer.samples.VerticalStackSample
 * @param modifier the modifier to apply to this layout.
 * @param state the state of the stack.
 * @param content a block that describes the content. Inside this block you can use methods like
 *   [StackScope.item] to add a single item or [StackScope.items] to add a collection of items.
 */
@Composable
public fun VerticalStack(
    modifier: Modifier = Modifier,
    state: StackState = rememberStackState(),
    content: StackScope.() -> Unit,
) {
    val latestContent = rememberUpdatedState(content)
    val stackItemHolderState =
        remember(state) {
            // Re-run the DSL to parse items only when the content lambda instance changes.
            derivedStateOf(referentialEqualityPolicy()) {
                    StackItemHolder(state, latestContent.value).also {
                        // Set the item count on the StackState immediately when the derived state
                        // re-evaluates (i.e., when content changes), even before recomposition.
                        state.itemCount = it.itemCount
                    }
                }
                .also {
                    // This second assignment is necessary to force the derivedStateOf lambda above
                    // (which is executed lazily) to execute synchronously inside this block.
                    state.itemCount = it.value.itemCount
                }
        }

    val singleItemScrollConstraintConnection =
        remember(state.pagerState) { SingleItemScrollConstraintConnection(state.pagerState) }

    VerticalPager(
        state = state.pagerState,
        modifier =
            modifier
                .then(StackInitialFocusElement(state))
                .stackScrim()
                .nestedScroll(singleItemScrollConstraintConnection),
        contentPadding = PaddingValues(bottom = RevealAreaSize),
        key = { page -> stackItemHolderState.value.getKey(page) },
        beyondViewportPageCount = MaxNextVisibleItemCount,
        flingBehavior =
            PagerDefaults.flingBehavior(
                state = state.pagerState,
                pagerSnapDistance = PagerSnapDistance.atMost(1),
                snapAnimationSpec = SnapAnimationSpec,
            ),
    ) { page ->
        val stackItemHolder = stackItemHolderState.value
        stackItemHolder.withInterval(page) { localIndex, itemInterval ->
            val key =
                itemInterval.getKeyOrDefault(globalIndex = page, localIntervalIndex = localIndex)
            val itemScope = itemInterval.getOrCreateItemScope(key)
            itemScope.index = page
            StackItemLayout(page = page, state = state, itemScope = itemScope) {
                itemInterval.item(itemScope, localIndex)
            }
        }
    }
}

@Composable
private fun StackItemLayout(
    page: Int,
    state: StackState,
    itemScope: StackItemScopeImpl,
    modifier: Modifier = Modifier,
    content: @Composable () -> Unit,
) {
    val focusRequester = remember { FocusRequester() }
    Layout(
        content = content,
        modifier =
            modifier
                .zIndex(-page.toFloat())
                .onPlaced { itemScope.coordinates = it }
                .maskItemsBelow(state, itemScope)
                .focusRequester(focusRequester)
                .onFocusChanged { state.onItemFocusChanged(page, it) },
    ) { measurables, constraints ->
        var maxWidth = 0
        var maxHeight = 0
        val placeables = measurables.fastMap {
            it.measure(constraints).also { placeable ->
                maxWidth = maxOf(maxWidth, placeable.width)
                maxHeight = maxOf(maxHeight, placeable.height)
            }
        }

        if (!isLookingAhead) {
            state.layoutInfoInternal.updateMeasuredHeight(index = page, height = maxHeight)
        }

        layout(maxWidth, maxHeight) {
            placeables.fastForEach { placeable ->
                placeable.placeRelativeWithLayer(x = 0, y = 0) {
                    val revealHeight = RevealAreaSize.roundToPx()
                    val topItem = state.topItem
                    when {
                        page.isTopItem(topItem = topItem) -> {
                            translationY =
                                state.topItemTranslationY(
                                    revealHeight = revealHeight,
                                    topItemHeight = placeable.height,
                                )

                            state.notifyAutoFocus(page, focusRequester)
                        }

                        page.isNextItem(topItem = topItem) -> {
                            translationY =
                                state.nextItemTranslationY(
                                    revealHeight = revealHeight,
                                    topItem = topItem,
                                    nextItemHeight = placeable.height,
                                )
                            val scale = state.nextItemScale()
                            scaleX = scale
                            scaleY = scale

                            state.notifyAutoFocus(page, focusRequester)
                        }

                        page.isNextNextItem(topItem = topItem) -> {
                            translationY =
                                state.nextNextItemTranslationY(
                                    revealHeight = revealHeight,
                                    topItem = topItem,
                                    nextNextItemHeight = placeable.height,
                                )
                            scaleX = NextItemMinScale
                            scaleY = NextItemMinScale
                        }
                    }
                }
            }
        }
    }
}

/**
 * Masks the items following the current item (items that are below the current item on Z-axis) if
 * the item's decoration fills the viewport width.
 */
private fun Modifier.maskItemsBelow(state: StackState, itemScope: StackItemScopeImpl): Modifier =
    this.drawWithContent {
        val viewportSize = state.layoutInfoInternal.viewportSize
        val viewportWidth = viewportSize.width.toFloat()

        if (itemScope.maskWidth >= (viewportWidth - DecorationWidthNoiseThresholdPx)) {
            // Only apply the mask to the items below (Z-axis) if the mask fills the viewport width.
            val viewportHeight = viewportSize.height.toFloat()
            drawRect(
                Color.Black,
                blendMode = BlendMode.DstOut,
                // The coordinate space here is for the item's layout, which changes position in the
                // stack viewport depending on the scroll position. We need to deduct the viewport
                // height to make the starting Y offset negative, so that the mask region extends
                // upwards to the top of the stack viewport.
                topLeft = Offset(x = 0f, y = itemScope.maskBottomY - viewportHeight),
                size = Size(width = viewportWidth, height = viewportHeight),
            )
        }

        drawContent()
    }

/**
 * The translation Y of the top item in pixels, which is always equal to the snapped position
 * offset.
 */
private fun StackState.topItemTranslationY(revealHeight: Int, topItemHeight: Int) =
    calculateTopPositionOffset(topItemHeight, revealHeight)

/** The current translation Y of the next item in pixels. */
private fun StackState.nextItemTranslationY(
    revealHeight: Int,
    topItem: Int,
    nextItemHeight: Int,
): Float {
    val nextPageOffset = pagerState.pageOffset(topItem + 1)
    val progress = topItemOffsetFraction
    val topItemHeight = layoutInfoInternal.measuredTopItemHeight

    val topPageTopPositionOffset = calculateTopPositionOffset(topItemHeight, revealHeight)
    val nextPageTopPositionOffset = calculateTopPositionOffset(nextItemHeight, revealHeight)

    val topItemBottom = topPageTopPositionOffset + topItemHeight
    // startOffset is the initial offset of the next item when it's revealed below the top item.
    val startOffset =
        calculateInitialBehindPosition(
            itemAboveBottom = topItemBottom + revealHeight,
            itemBehindHeight = nextItemHeight,
        )

    val currentOffset =
        lerp(start = startOffset, stop = nextPageTopPositionOffset, fraction = progress)

    return currentOffset - nextPageOffset
}

/** The current scale of the next item. */
private fun StackState.nextItemScale(): Float {
    val progress = topItemOffsetFraction
    return lerp(start = NextItemMinScale, stop = 1f, fraction = progress)
}

/** The current translation Y of the next-next item in pixels. */
private fun StackState.nextNextItemTranslationY(
    revealHeight: Int,
    topItem: Int,
    nextNextItemHeight: Int,
): Float {
    val nextNextPageOffset = pagerState.pageOffset(topItem + 2)
    val nextItemHeight = layoutInfoInternal.measuredNextItemHeight

    // Calculate where the next item *would* be if it were currently the top item.
    val nextItemTopPositionOffset = calculateTopPositionOffset(nextItemHeight, revealHeight)
    val nextItemTopPositionBottom = nextItemTopPositionOffset + nextItemHeight

    // Calculate the static target position the next-next item.
    // We add revealHeight to nextItemBottom to ensure the item waits at the exact position
    // where it will need to be when it transitions to being the next item.
    val offset =
        calculateInitialBehindPosition(
            itemAboveBottom = nextItemTopPositionBottom + revealHeight,
            itemBehindHeight = nextNextItemHeight,
        )

    return offset - nextNextPageOffset
}

/**
 * Calculates the offset from the top of the viewport for an item of a given height for the item's
 * top snapped position. Items are aligned to the bottom of the stack layout.
 */
private fun StackState.calculateTopPositionOffset(itemHeight: Int, revealHeight: Int): Float {
    val viewportHeight = layoutInfoInternal.viewportSize.height
    return (viewportHeight - itemHeight - revealHeight).coerceAtLeast(0).toFloat()
}

/** Calculates the initial offset for an item when it is positioned behind the item above it. */
private fun calculateInitialBehindPosition(itemAboveBottom: Float, itemBehindHeight: Int): Float =
    itemAboveBottom - itemBehindHeight * NextItemPositioningScale

/**
 * The main axis offset of the item in pixels from the top of the viewport.
 *
 * If the page is in the visible list (inside the viewport), we return its exact offset. If it's not
 * in the list but is composed (due to beyondViewportPageCount), we calculate where it *should* be
 * based on the current page and offset fraction.
 *
 * Note: this method assumes that the Pager's layoutInfo is already available.
 */
private fun PagerState.pageOffset(page: Int): Int {
    val layoutInfo = layoutInfo

    // First check if the page is already visible.
    val visiblePage = layoutInfo.visiblePagesInfo.fastFirstOrNull { it.index == page }
    if (visiblePage != null) return visiblePage.offset

    // Calculate the distance in "pages" from the current snap position.
    // (page - currentPage) gives the index distance.
    // (- currentPageOffsetFraction) accounts for the sub-page scroll.
    val offsetFromCurrentPage = page - currentPage - currentPageOffsetFraction
    val stride = layoutInfo.pageSize + layoutInfo.pageSpacing
    return (offsetFromCurrentPage * stride).roundToInt()
}

/** The size of the area where the items beneath the top of the stack item are revealed. */
internal val RevealAreaSize = 18.dp

/** The maximum number of items that can be visible at the same time in addition to the top item. */
private const val MaxNextVisibleItemCount = 2

/**
 * The scale of the next item when it is fully behind the top item, and the default scale of the
 * item behind it (the next-next item).
 */
private const val NextItemMinScale = 0.94f

/** The scale factor that's between 1.0 and [NextItemMinScale], which is used in positioning. */
private const val NextItemPositioningScale = 0.97f // (1f + NextItemMinScale) / 2f

/** The animation spec used for snapping stack items. */
private val SnapAnimationSpec =
    spring(
        dampingRatio = 0.56f,
        stiffness = 118f,
        visibilityThreshold = Int.VisibilityThreshold.toFloat(),
    )

<br />

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Security

No issues found by our static scan· No issues found

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Attribution and licence

Created by
Google LLC
Publisher
Android (Google) (Official)
Distributed by
Android (Google)
Source
android/skills · xr/display-glasses-with-jetpack-compose-glimmer · commit bf68e2d
Licence
Apache License 2.0 (redistribution allowed)
  • LICENSE (added by AutoGineer from LICENSE.txt)
  • ATTRIBUTION.md (written by AutoGineer)
Synced

AutoGineer is not affiliated with or endorsed by Google LLC or Android (Google).

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download display-glasses-with-jetpack-compose-glimmer.zip