What problem does it solve?
This Skill helps you identify valuable in-app user workflows and expose them as AppFunctions so LLM agents can discover and execute them on-device without relying on UI navigation.
Core Features & Use Cases
- Workflow Discovery for AppFunctions: Analyze an app’s codebase to recommend high-value AppFunctions (e.g., creating notes, playing media, sending automated/AI-triggered messages, voice commands, and system shortcuts).
- Implementation & System Configuration: Generate Kotlin AppFunction implementations with the required Gradle/KSP setup and app metadata configuration so the system can index and surface functions.
- KDoc Refinement for Agent Compatibility: Refine function and serializable documentation to produce agent-usable descriptions (including MCP alignment) so the model invokes tools correctly.
- Testing & Debugging via ADB: Provide commands to list, invoke, enable/disable, and validate AppFunctions on a connected device or emulator.
Quick Start
Ask an AI to analyze your Android app for AppFunctions by targeting workflows you want agents to perform (like sending a message or creating a note), then generate the required Kotlin implementations and ADB test steps so you can verify registration on a device.