Meta Wearables Device Access Toolkit for Android

The Meta Wearables Device Access Toolkit enables developers to utilize Meta’s AI glasses to build hands-free wearable experiences into their mobile applications. By integrating this SDK, developers can reliably connect to Meta’s AI glasses and leverage capabilities like video and audio streaming and photo capture.

The Wearables Device Access Toolkit is in developer preview. Developers can access our SDK and documentation, test on supported AI glasses, and create organizations and release channels to share with test users.

Documentation & Community

Find our full developer documentation on the Wearables Developer Center.

You can find an overview of the Wearables Developer Center here. Create an account to stay informed of all updates, report bugs and register your organization. Set up a project and release channel to share your integration with test users.

For help, discussion about best practices or to suggest feature ideas visit our discussions forum.

See the changelog for the latest updates.

Including the SDK in your project

The Wearables Device Access Toolkit is published to Maven Central under the com.meta.wearable group, so no access token or extra repository definition is required. The mavenCentral() repository already declared in your settings.gradle.kts is all you need.

1. Declare the Wearables Device Access Toolkit artifacts in libs.versions.toml

Check the available versions on the artifact page.

[versions]
mwdat = "1.0.1"

[libraries]
mwdat-core = { group = "com.meta.wearable", name = "mwdat-core", version.ref = "mwdat" }
mwdat-camera = { group = "com.meta.wearable", name = "mwdat-camera", version.ref = "mwdat" }
mwdat-display = { group = "com.meta.wearable", name = "mwdat-display", version.ref = "mwdat" }
mwdat-inputs = { group = "com.meta.wearable", name = "mwdat-inputs", version.ref = "mwdat" }
mwdat-motion = { group = "com.meta.wearable", name = "mwdat-motion", version.ref = "mwdat" }
mwdat-mockdevice = { group = "com.meta.wearable", name = "mwdat-mockdevice", version.ref = "mwdat" }
mwdat-speech = { group = "com.meta.wearable", name = "mwdat-speech", version.ref = "mwdat" }

2. Add the required components as dependencies in your app’s build.gradle.kts

dependencies {
    implementation(libs.mwdat.core)
    implementation(libs.mwdat.camera)
    implementation(libs.mwdat.display)
    implementation(libs.mwdat.inputs)
    implementation(libs.mwdat.motion)
    implementation(libs.mwdat.mockdevice)
    implementation(libs.mwdat.speech)
}

Developer Terms

• By using the Wearables Device Access Toolkit, you agree to our Meta Wearables Developer Terms, including our Acceptable Use Policy.
• By enabling Meta integrations, including through this SDK, Meta may collect information about how users’ Meta devices communicate with your app. Meta will use this information collected in accordance with our Privacy Policy.
• You may limit Meta’s access to data from users’ devices by following the instructions below.

Opting out of data collection

To configure analytics settings in your Meta Wearables DAT Android app, add the following <meta-data> element to your app’s AndroidManifest.xml file within the <application> element:

<meta-data
    android:name="com.meta.wearable.mwdat.ANALYTICS_OPT_OUT"
    android:value="true"
    />

Default behavior: If the ANALYTICS_OPT_OUT metadata is missing or set to false, analytics are enabled (i.e., you are not opting out). Set to true to disable data collection.

Note: In other words, this setting controls whether or not you’re opting out of analytics:

• true = Opt out (analytics disabled)
• false = Opt in (analytics enabled)

Complete example:

<application
    android:name=".MyApplication"
    android:label="MyApp"
    android:icon="@mipmap/app_launcher">

    <!-- Required: Your application ID from Wearables Developer Center -->
    <meta-data
        android:name="com.meta.wearable.mwdat.APPLICATION_ID"
        android:value="your_app_id_here"
        />

    <!-- Optional: Disable analytics -->
    <meta-data
        android:name="com.meta.wearable.mwdat.ANALYTICS_OPT_OUT"
        android:value="true"
        />

    <!-- Your activities and other components -->
</application>

Crash reporting

The Wearables Device Access Toolkit can capture crashes originating from SDK code, store them locally, and chain with any existing uncaught-exception handler your app installs. Crash reporting is enabled by default.

To opt out, add the following <meta-data> element to your app’s AndroidManifest.xml file within the <application> element:

<meta-data
    android:name="com.meta.wearable.mwdat.CRASH_REPORTING_OPT_OUT"
    android:value="true"
    />

Default behavior: If the CRASH_REPORTING_OPT_OUT metadata is missing or set to false, crash reporting is enabled. Set it to true to disable SDK crash capture.

AI-Assisted Development

This repository ships one public DAT knowledge base as a single plugin, plus file-based artifacts for the tools that do not use plugins:

Tool	Public artifact	Recommended setup
Muse Code, Claude Code, Codex	plugins/mwdat-android/	Add this GitHub repo as a marketplace, then install mwdat-android
GitHub Copilot	.github/copilot-instructions.md	Auto-loaded by Copilot in VS Code
Cursor	.cursor/rules/*.mdc	Auto-loaded with glob-based triggers
AGENTS.md-compatible tools	AGENTS.md	Portable fallback for agents that read AGENTS.md
MCP-compatible editors	https://mcp.developer.meta.com/wearables	Connect as a remote HTTP MCP server; no authentication required

Muse Code, Claude Code, and Codex install from the plugin payload under plugins/. Copilot, Cursor, and AGENTS.md readers use the native file-based artifacts at repo root.

Muse Code

muse plugins marketplace add mwdat-android-marketplace https://github.com/facebook/meta-wearables-dat-android
muse plugins install mwdat-android@mwdat-android-marketplace
muse plugins approve mwdat-android

Or use the helper script:

./install-skills.sh muse

Muse Code also reads AGENTS.md from your project root with no setup at all.

Claude Code

claude plugin marketplace add facebook/meta-wearables-dat-android
claude plugin install mwdat-android@mwdat-android-marketplace

Or use the helper script:

./install-skills.sh claude

Codex

codex plugin marketplace add facebook/meta-wearables-dat-android
codex plugin add mwdat-android@mwdat-android-marketplace

Or use the helper script:

./install-skills.sh codex

Other tool installs

Use the installer when you want the repo-native file surfaces for other tools:

./install-skills.sh copilot   # .github/copilot-instructions.md
./install-skills.sh cursor    # .cursor/rules/*.mdc
./install-skills.sh agents    # AGENTS.md
./install-skills.sh all       # Muse/Claude/Codex when available, plus Copilot/Cursor/AGENTS.md

Or run the helper remotely:

curl -sL https://raw.githubusercontent.com/facebook/meta-wearables-dat-android/main/install-skills.sh | bash

What’s included

• Getting started — SDK setup, Gradle integration, manifest configuration
• Camera streaming — Session and Stream setup, video frames, and in-stream photo capture
• Audio streaming — Receive synchronized PCM audio through Camera Stream (experimental; unavailable for production publishing)
• Camera capture — Capture standalone high-quality photos with progress and typed errors (experimental; unavailable for production publishing)
• Display — Use Display features on the Meta Ray-Ban Display glasses
• Inputs — Receive navigation, button, capture, and drag interactions from glasses
• Motion — Stream accelerometer, gyroscope, magnetometer, and orientation samples with explicit lifecycle control
• Speech — Recognize on-device speech, handle partial/final results, and test with MockDeviceKit
• Voice Invocations — Launch or activate an app from Hey Meta and acknowledge each action
• MockDevice testing — Test without physical glasses using MockDeviceKit
• DAT docs MCP — Connect the hosted docs MCP server and search live DAT documentation with search_dat_docs
• Live debugging MCP — Read-only runtime diagnosis from app-visible DAT debug events
• Session lifecycle — Session and stream state, pause/resume behavior, and device availability monitoring
• Permissions & registration — App registration with Meta AI and device permission flows
• Debugging — Common issues, Developer Mode, version compatibility, and session and stream diagnosis
• Sample app guide — Building a complete DAT app

For static reference context, point your AI tool at the llms.txt endpoint. Installing the plugin registers https://mcp.developer.meta.com/wearables for live documentation search in Muse Code, Claude Code, and Codex. Other MCP-compatible editors can connect to that endpoint directly and use search_dat_docs. The public docs MCP server does not require authentication; do not configure tokens, OAuth, or custom authorization headers for it.

License

See the LICENSE file.