Skip to content

Latest commit

 

History

History
82 lines (59 loc) · 3.54 KB

ROADMAP.md

File metadata and controls

82 lines (59 loc) · 3.54 KB

🗺 Roadmap

MobileGraph is being developed in incremental phases, with each phase building on a stable, modular, and extensible architecture.

Our mission is to become the Kotlin-first AI framework for the Android ecosystem and Kotlin Multiplatform including IOS, enabling developers to build everything from simple AI assistants to enterprise-grade, autonomous agentic applications.

Phase Status Focus
✅ Phase 1 Completed Core Runtime & Provider Abstraction
✅ Phase 2 Completed Knowledge Layer (RAG)
✅ Phase 3 Completed Agent Runtime (Durable Workflows)
✅ Phase 4 Completed Multi-Model Cloud Ecosystem
✅ Phase 5 Completed Model Context Protocol (MCP)
✅ Phase 6 Completed Agent-to-Agent (A2A) Protocol
🔜 Phase 7 Planned Local AI & Edge Inference
🔮 Phase 8 Future Android & Automotive Ecosystem
🚀 Phase 9 Vision MobileGraph Studio (Visual Builder)

✅ Phase 1 — Core Runtime

Foundation for AI-powered applications. Abstraction layer for chat models, session management, middleware, and structured outputs.

✅ Phase 2 — Knowledge Layer (RAG)

Enable Retrieval-Augmented Generation. Includes document loaders, text splitters, vector store abstractions, and a retrieval DSL.

✅ Phase 3 — Agent Runtime

Durable, mobile-first agent execution engine. Support for stateful graphs, autonomous tool use, Human-in-the-loop (HITL), and automatic process death recovery.

✅ Phase 4 — Model Ecosystem

Expanded cloud provider suite. Native integration for OpenAI, Gemini, Claude, DeepSeek, Hugging Face, and OpenRouter. Includes multi-modal vision and intelligent model routing.

✅ Phase 5 — Model Context Protocol (MCP)

Enable seamless interoperability with external tools and services through MCP. Includes MCP client, Streamable HTTP and SSE transports, and dynamic tool discovery.

✅ Phase 6 — Agent-to-Agent (A2A) Protocol

Enable multi-agent collaboration across heterogeneous systems via A2A Protocol v1.0.

Highlights

  • A2A Client for Android & KMP (:intelligence:mobilegraph-a2a)
  • Agent Card discovery (/.well-known/agent-card.json)
  • Task delegation via JSON-RPC (SendMessage, GetTask, CancelTask)
  • Real-time SSE streaming updates (SendStreamingMessage)
  • Human-in-the-loop (HITL) task handling (input-required)
  • Multi-modal artifacts and conversation history tracking
  • Automatic protocol version fallback (v1.0 → v0.3)

🔜 Phase 7 — Local AI & Edge Inference

Bring private, offline AI directly to Android devices.

Planned Features

  • On-device Small Language Model (SLM) inference
  • llama.cpp & GGUF integration
  • ONNX Runtime & MediaPipe GenAI support
  • Qualcomm AI Engine / NPU acceleration
  • Hybrid local + cloud routing

🔮 Phase 8 — Android Ecosystem

Extend MobileGraph across the Android platform family.

Highlights

  • Android: Jetpack Compose integration, WorkManager AI tasks.
  • Automotive: IVI AI framework, Vehicle Data integration, distraction-aware execution.
  • TV/XR: Voice-first interactions and spatial AI agents.

🚀 Phase 9 — MobileGraph Studio

Create a visual development environment for designing AI workflows.

Highlights

  • Drag-and-drop workflow builder
  • Visual State Graph editor & debugger
  • Prompt Playground & RAG Inspector
  • Token & cost optimization dashboard