MCP for mobile apps.
Give your Android app agentic powers.
Apps describe what they can do. AI Assistant (Hark) discovers and invokes those capabilities. On-device. Open protocol.
Hark means "to listen." It's also short for my name "Harkirat".
Start in the right place
OACP
The open standard. Apps ship oacp.json. Assistants discover via ContentProvider.
Read the protocol →Hark
Open-source voice assistant. On-device AI with EmbeddingGemma + Qwen3. Wake word, overlay UI, async results.
Meet Hark →SDK
Easily add OACP to any Android app with these SDKs.
Get the SDK →How it works
- 1Describe the appThe app ships oacp.json and OACP.md so an assistant can understand the actions, parameters, and vocabulary.
- 2Discover capabilitiesAn OACP-compatible assistant scans the device for .oacp ContentProviders and reads each manifest at runtime.
- 3Match the requestThe assistant uses its on-device pipeline to match the user's request to the best capability and extract any parameters.
- 4Invoke the appThe assistant dispatches an Android intent. Background tasks use broadcasts and UI flows use activities.
- 5Return the resultIf the action is async, the app sends a structured result back and the assistant can speak or display it.
Background action (round trip)
"Hey Hark, what's the weather?"
│
▼
┌─────────────┐ broadcast ┌─────────────────┐
│ Hark │─────────────►│ Weather App │
│ on-device │ (OACP) │ │
│ AI │◄─────────────│ fetches data │
└─────────────┘ ACTION_RESULT└─────────────────┘
│
▼
"Currently 22°, partly cloudy"Foreground action (one way)
"Hey Hark, take a picture with
front camera in 2 seconds"
│
▼
┌─────────────┐ (OACP) ┌─────────────────┐
│ Hark │─────────────►│ Camera App │
│ on-device │ activity │ │
│ AI │ │ opens camera, │
└─────────────┘ │ 2s countdown │
└─────────────────┘Ecosystem
8 apps and counting. Breezy Weather, Binary Eye, Wikipedia, and more. Each one ships an oacp.json and works with any OACP assistant out of the box.
See all appsWhat's next
Wake word detection is shipped. Background listening, self-hosted inference, and disambiguation UI are next. Everything is open source.
See the full roadmap