Edge To Edge logo

Edge To Edge

OrganizationPopular
android
edge-to-edge

Use this skill to migrate your Jetpack Compose app to add adaptive edge-to-edge support and troubleshoot common issues. Use this skill to fix UI components (like buttons or lists) that are obscured by or overlapping with the navigation bar or status bar, fix IME insets, and fix system bar legibility.

Overview

Publisherandroid
Repositoryskills
Skill nameedge-to-edge
Stars
7.4K
Forks
484
Bundled files
Instructions only
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by android on GitHub. Read the source before you install it.

Installation

Install the Edge To Edge AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/android/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/system/edge-to-edge .claude/skills/edge-to-edge
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Edge To Edge in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Edge To Edge on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Edge To Edge is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Prerequisites

  • Project MUST use Android Jetpack Compose.
  • Project MUST target SDK 35 or later. If the SDK is lower than 35, increase the SDK to 35.

Step 1: plan

  1. Locate and analyze all Activity classes to detect which have existing edge-to-edge support. For every Activity without edge-to-edge, plan to make each Activity edge-to-edge.
  2. In each Activity, Locate and analyze all lists and FAB components to detect which have existing edge-to-edge support. For every component without edge-to-edge support, plan to make each of these components edge-to-edge.
  3. In each Activity, scan for TextField, OutlinedTextField, or BasicTextField. If found, then you MUST verify the IME doesn't hide the input field by following the IME section of this skill.

Step 2: add edge-to-edge support

  1. Add enableEdgeToEdge before setContent in onCreate in each Activity that does not already call enableEdgeToEdge.
  2. Add android:windowSoftInputMode="adjustResize" in the AndroidManifest.xml for all Activities that use a soft keyboard.

Step 3: apply insets

  • The app MUST apply system insets, or align content to rulers, so critical UI remains tappable. Choose only one method to avoid double padding:

    1. PREFERRED: When available, use Scaffolds and pass PaddingValues to the content lambda.
    kotlin
    Scaffold { innerPadding ->
        // innerPadding accounts for system bars and any Scaffold components
        LazyColumn(
            modifier = Modifier
                .fillMaxSize()
                .consumeWindowInsets(innerPadding),
            contentPadding = innerPadding
        ) { /* Content */ }
    }
    1. PREFERRED: When available, use the automatic inset handling or padding modifiers in material components.

      • Material 3 Components manages safe areas for its own components, including:
        • TopAppBar
        • SmallTopAppBar
        • CenterAlignedTopAppBar
        • MediumTopAppBar
        • LargeTopAppBar
        • BottomAppBar
        • ModalDrawerSheet
        • DismissibleDrawerSheet
        • PermanentDrawerSheet
        • ModalBottomSheet
        • NavigationBar
        • NavigationRail
      • For Material 2 Components, use the windowInsetsparameter to apply insets manually for BottomAppBar, TopAppBar and BottomNavigation. DO NOT apply padding to the parent container; instead, pass insets directly to the App Bar component. Applying padding to the parent container prevents the App Bar background from drawing into the system bar area. For example, for TopAppBar, choose only one of the following options:
        1. PREFERRED: TopAppBar(windowInsets = AppBarDefaults.topAppBarWindowInsets)
        2. TopAppBar(windowInsets = WindowInsets.systemBars.exclude(WindowInsets.navigationBars))
        3. TopAppBar(windowInsets = WindowInsets.systemBars.add(WindowInsets.captionBar))
    2. For components outside a Scaffold, use padding modifiers, such as Modifier.safeDrawingPadding() or Modifier.windowInsetsPadding(WindowInsets.safeDrawing).

      kotlin
      Box(
          modifier = Modifier
              .fillMaxSize()
              .safeDrawingPadding()
      ) {
          Button(
              onClick = {},
              modifier = Modifier.align(Alignment.BottomCenter)
          ) {
              Text("Login")
          }
      }
    3. For deeply nested components with excessive padding, use WindowInsetsRulers (e.g. Modifier.fitInside(WindowInsetsRulers.SafeDrawing.current)). See the IME section for a code sample.

    4. When you need an element (e.g. a custom header or decorative scrim) to equal the dimensions of a system bar, use inset size modifiers (e.g. Modifier.windowInsetsTopHeight(WindowInsets.systemBars)). See the Lists section for a code sample.

Adaptive Scaffolds

  • NavigationSuiteScaffold manages safe areas for its own components, like the NavigationRail or NavigationBar. However, the adaptive scaffolds (e.g. NavigationSuiteScaffold, ListDetailPaneScaffold) don't propagate PaddingValues to their inner contents. You MUST apply insets to individual screens or components (e.g., list contentPadding or FAB padding) as described in Step 3 . DO NOT apply safeDrawingPadding or similar modifiers to the NavigationSuiteScaffold parent. This clips and prevents an edge-to-edge screen.

IME

  • For each Activity with a soft keyboard, check that android:windowSoftInputMode="adjustResize" is set in the AndroidManifest.xml. DO NOT use SOFT_INPUT_ADJUST_RESIZE because it is deprecated. Then, maintain focus on the input field. Choose one:
      1. PREFERRED: Add Modifier.fitInside(WindowInsetsRulers.Ime.current) to the content container. This is preferred over imePadding() because it reduces jank and extra padding caused by forgetting to consume insets upstream in the hierarchy.
      1. Add imePadding to the content container. The padding modifier MUST be placed before Modifier.verticalScroll(). Do NOT use Modifier.imePadding() if the parent already accounts for the IME with contentWindowInsets (e.g. contentWindowInsets = WindowInsets.safeDrawing). Doing so will cause double padding.

IMEs with Scaffolds code patterns

RIGHT

RIGHT because contentWindowInsets contains IME insets, which are passed to the content lambda as innerPadding.

kotlin
// RIGHT
Scaffold(contentWindowInsets = WindowInsets.safeDrawing) { innerPadding ->
    Column(
        modifier = Modifier
            .padding(innerPadding)
            .consumeWindowInsets(innerPadding)
            .verticalScroll(rememberScrollState())
    ) { /* Content */ }
}

RIGHT because fitInside fits the content to the IME insets regardless of contentWindowInsets.

kotlin
// RIGHT
Scaffold() { innerPadding ->
    Column(
        modifier = Modifier
            .padding(innerPadding)
            .consumeWindowInsets(innerPadding)
            .fitInside(WindowInsetsRulers.Ime.current)
            .verticalScroll(rememberScrollState())
    ) { /* Content */ }
}

RIGHT because the default contentWindowInsets does not contain IME insets, and imePadding() applies IME insets:

kotlin
// RIGHT
Scaffold() { innerPadding ->
    Column(
        modifier = Modifier
            .padding(innerPadding)
            .consumeWindowInsets(innerPadding)
            .imePadding()
            .verticalScroll(rememberScrollState())
    ) { /* Content */ }
}
WRONG

WRONG because there will be excess padding when the IME opens. IME insets are applied twice, once with innerPadding, which contains IME insets from the passed contentWindowInsets values, and once with imePadding:

kotlin
// WRONG
Scaffold(contentWindowInsets = WindowInsets.safeDrawing) { innerPadding ->
    Column(
        modifier = Modifier
            .padding(innerPadding)
            .imePadding()
            .verticalScroll(rememberScrollState())
    ) { /* Content */ }
}

WRONG because the IME will cover up the content. Scaffold's default contentWindowInsets does NOT contain IME insets.

kotlin
// WRONG
Scaffold() { innerPadding ->
    Column(
        modifier = Modifier
            .padding(innerPadding)
            .verticalScroll(rememberScrollState())
    ) { /* Content */ }
}

IMEs without Scaffolds code patterns

RIGHT

The following code samples WILL NOT cause excessive padding.

kotlin
// RIGHT
Box(
    // Insets consumed
    modifier = Modifier.safeDrawingPadding() // or imePadding(), safeContentPadding(), safeGesturesPadding()
) {
    Column(
        modifier = Modifier.imePadding()
    ) { /* Content */ }
}

kotlin
// RIGHT
Box(
    // Insets consumed
    modifier = Modifier.windowInsetsPadding(WindowInsets.safeDrawing) // or WindowInsets.ime, WindowInsets.safeContent, WindowInsets.safeGestures
) {
    Column(
        modifier = Modifier.imePadding()
    ) { /* Content */ }
}

kotlin
// RIGHT
Box(
    // Insets not consumed, but irrelevant due to fitInside
    modifier = Modifier.padding(WindowInsets.safeDrawing.asPaddingValues()) // or WindowInsets.ime.asPaddingValues(), WindowInsets.safeContent.asPaddingValues(), WindowInsets.safeGestures.asPaddingValues()
) {
    Column(
        modifier = Modifier
            .fillMaxSize()
            .fitInside(WindowInsetsRulers.Ime.current)
    ) { /* Content */ }
}
WRONG

The following code sample WILL cause excessive padding because IME insets are applied twice:

kotlin
// WRONG
Box(
    // Insets not consumed
    modifier = Modifier.padding(WindowInsets.safeDrawing.asPaddingValues()) // or WindowInsets.ime.asPaddingValues(), WindowInsets.safeContent.asPaddingValues(), WindowInsets.safeGestures.asPaddingValues()
) {
    Column(
        modifier = Modifier.imePadding()
    ) { /* Content */ }
}

Navigation Bar Contrast & System Bar Icons

  • If the Activity uses enableEdgeToEdge from WindowCompat, you MUST set isAppearanceLightNavigationBars and isAppearanceLightStatusBars to the inverse of the device theme for apps that support light and dark theme so the system bar icons are legible. It's recommended to do this in your theme file. DO NOT do this if the Activities use enableEdgeToEdge from ComponentActivity because it handles the icon colors automatically.

    kotlin
    // Only use if calling `enableEdgeToEdge` from `WindowCompat`.
    // Apply to your theme file.
    @Composable
    fun MyTheme(
        darkTheme: Boolean = isSystemInDarkTheme(),
        content: @Composable () -> Unit
    ) {
        val view = LocalView.current
        if (!view.isInEditMode) {
            SideEffect {
                val window = (view.context as? Activity)?.window ?: return@SideEffect
                val controller = WindowCompat.getInsetsController(window, view)
    
                // Dark icons for Light Mode (!darkTheme), Light icons for Dark Mode
                controller.isAppearanceLightStatusBars = !darkTheme
                controller.isAppearanceLightNavigationBars = !darkTheme
            }
        }
    
        MaterialTheme(content = content)
    }
  • If any screen uses a Scaffold or a NavigationSuiteScaffold with a bottom bar (e.g., BottomAppBar, NavigationBar), set window.isNavigationBarContrastEnforced = false in the corresponding Activity for SDK 29+. This prevents the system from adding a translucent background to the navigation bar, verifying your bottom bar colors extend to the bottom of the screen.

Lists

  • Apply inset padding (like Scaffold's innerPadding) to the contentPadding parameter of scrollable components (e.g. LazyColumn, LazyRow). DO NOT apply it as a Modifier.padding() to the list's parent container, as this clips the content and prevents it from scrolling behind the system bars.
  • Create a translucent composable covering the system bar so that the icons are still legible.
kotlin
class SystemBarProtectionSnippets : ComponentActivity() {

    override fun onCreate(savedInstanceState: Bundle?) {
        super.onCreate(savedInstanceState)

        // enableEdgeToEdge sets window.isNavigationBarContrastEnforced = true
        // which is used to add a translucent scrim to three-button navigation
        enableEdgeToEdge()

        setContent {
            MyTheme {
                // Main content
                MyContent()

                // After drawing main content, draw status bar protection
                StatusBarProtection()
            }
        }
    }
}

@Composable
private fun StatusBarProtection(
    color: Color = MaterialTheme.colorScheme.surfaceContainer,
) {
    Spacer(
        modifier = Modifier
            .fillMaxWidth()
            .height(
                with(LocalDensity.current) {
                    (WindowInsets.statusBars.getTop(this) * 1.2f).toDp()
                }
            )
            .background(
                brush = Brush.verticalGradient(
                    colors = listOf(
                        color.copy(alpha = 1f),
                        color.copy(alpha = 0.8f),
                        Color.Transparent
                    )
                )
            )
    )
}

Dialogs

If both the following conditions are true, then the Dialog is full screen and must be made edge-to-edge:

  1. The DialogProperties contains usePlatformDefaultWidth = false.
  2. The Dialog calls Modifier.fillMaxSize().

To make a full screen Dialog edge-to-edge, set decorFitsSystemWindows = false in the DialogProperties.

kotlin
Dialog(
    onDismissRequest = { /* Handle dismiss */ },
    properties = DialogProperties(
        // 1. Allows the dialog to span the full width of the screen
        usePlatformDefaultWidth = false,
        // 2. Allows the dialog to draw behind status and navigation bars
        decorFitsSystemWindows = false
    )
) { /* Content */ }

Checklist

  • [ ] Does every Activity call enableEdgeToEdge()?
  • [ ] Is adjustResize set in the AndroidManifest.xml?
  • [ ] Does every TextField, OutlinedTextField, or BasicTextField have a parent with imePadding(), fitInside, Modifier.safeDrawingPadding(), Modifier.safeContentPadding(), Modifier.safeGesturesPadding(), or contentWindowInsets set to WindowInsets.safeDrawing or WindowInsets.ime?
  • [] Does the first and last list item draw away from the system bars by passing insets to contentPadding?
  • [] Do FABs draw above the navigation bars by either being inside a Scaffold or by applying Modifier.safeDrawingPadding()?
  • [] Does the project build? Run ./gradlew build to be sure.

Frequently asked questions

What does the Edge To Edge AI skill do?

Use this skill to migrate your Jetpack Compose app to add adaptive edge-to-edge support and troubleshoot common issues. Use this skill to fix UI components (like buttons or lists) that are obscured by or overlapping with the navigation bar or status bar, fix IME insets, and fix system bar legibility.

Why use Edge To Edge on TypingMind?

Because you install it once and use it with any model. Edge To Edge is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Edge To Edge in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/android/skills/tree/main/system/edge-to-edge. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Edge To Edge?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Edge To Edge?

As many as you like. As long as a model supports skills, you can use Edge To Edge with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Edge To Edge AI skill free?

Yes. It is published on GitHub by android under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇