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 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.
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.
You can add the Wearables Device Access Toolkit to your Gradle project by following the steps below.
You will need to provide a personal access token (classic) with at least read:packages scope as an environment variable named GITHUB_TOKEN or
by adding it as a property named github_token in your local.properties file.
See SDK for Android setup for more details.
val localProperties =
Properties().apply {
val localPropertiesPath = rootDir.toPath() / "local.properties"
if (localPropertiesPath.exists()) {
load(localPropertiesPath.inputStream())
}
}
dependencyResolutionManagement {
...
repositories {
...
maven {
url = uri("https://maven.pkg.github.com/facebook/meta-wearables-dat-android")
credentials {
username = "" // not needed
password = System.getenv("GITHUB_TOKEN") ?: localProperties.getProperty("github_token")
}
}
}
}Check the available versions in GitHub Packages.
[versions]
mwdat = "0.7.0"
[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-mockdevice = { group = "com.meta.wearable", name = "mwdat-mockdevice", version.ref = "mwdat" }dependencies {
implementation(libs.mwdat.core)
implementation(libs.mwdat.camera)
implementation(libs.mwdat.mockdevice)
}- 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.
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>This repository ships one public DAT knowledge base in two first-class formats:
| Tool | Public artifact | Recommended setup |
|---|---|---|
| Claude Code | .claude-plugin/marketplace.json + plugins/mwdat-android/.claude-plugin/plugin.json |
Add this GitHub repo as a marketplace, then install mwdat-android |
| Codex | plugins/mwdat-android/.codex-plugin/plugin.json |
Install the plugin from a cloned checkout of this repo |
| 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.facebook.com/wearables_dat |
Connect as a remote HTTP MCP server |
Claude and Codex install from the plugin payload under plugins/. Copilot, Cursor, and AGENTS.md readers use the native file-based artifacts at repo root.
claude plugin marketplace add facebook/meta-wearables-dat-android
claude plugin install mwdat-android@mwdat-android-marketplaceOr use the helper script:
./install-skills.sh claudegit clone https://github.com/facebook/meta-wearables-dat-android.git
cd meta-wearables-dat-android
codex plugin install ./plugins/mwdat-androidOr use the helper script:
./install-skills.sh codexUse 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 # Claude/Codex when available, plus Copilot/Cursor/AGENTS.mdOr run the helper remotely:
curl -sL https://raw.githubusercontent.com/facebook/meta-wearables-dat-android/main/install-skills.sh | bash- Getting started — SDK setup, Gradle integration, manifest configuration
- Camera streaming — Session and Stream capability setup, video frames, resolution/frame rate, photo capture
- MockDevice testing — Test without physical glasses using MockDeviceKit
- 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. For live documentation search in MCP-compatible editors, connect https://mcp.facebook.com/wearables_dat and use search_dat_docs.
See the LICENSE file.