Life Agent RU YEE (LARY) is an AI-driven framework for autonomous life management. It targets users seeking automated daily routines, handling tasks like meal planning, grocery shopping, and household automation through a multi-agent system and natural language interaction. The core benefit is offloading routine management to an intelligent, adaptable agent.

How It Works

LARY employs a pluggable, multi-agent architecture orchestrated by a FastAPI backend. Core components include an intent router, agent communication module, and a plugin registry for agents, memory, and extensions. It leverages SSE for real-time frontend updates on tool execution and agent states. A key differentiator is its direct Android device automation via uiautomator2, enabling programmatic interaction with apps like Hema for automated grocery purchasing, bridging AI planning with real-world actions.

Quick Start & Requirements

Prerequisites include Python 3.11+, Node.js 18+, PostgreSQL 16, and Redis 7. Installation involves cloning the repository, configuring LLM API keys in .env (Volcengine, OpenAI, Anthropic, or Ollama fallback), and starting infrastructure via docker compose up -d postgres redis. The backend is launched with pip install -e ".[dev]" && python main.py, and the frontend with cd web && npm install && npm run dev. A Docker one-click deployment is available via docker compose up --build -d. Development mode offers volume mounting and USB passthrough.

Highlighted Details

• Multi-Agent Collaboration: Dedicated agents for meal planning and purchasing autonomously coordinate tasks from recipe selection to adding items to a shopping cart.
• Pluggable Architecture: Agents, memory modules, and extensions are designed as hot-loadable plugins, supporting community development and customization.
• Real-time SSE: Server-Sent Events provide live streaming of tool call progress, agent thinking states, and operational logs to the user interface.
• Android Automation: Integrates uiautomator2 for direct control of Android devices, automating interactions within specific applications for tasks like online shopping.
• User Profiling: Dynamically builds user profiles based on household size and dietary preferences to personalize recommendations.
• SkillHub: An upcoming marketplace for discovering, installing, and managing community-developed plugins.

Maintenance & Community

The README does not detail specific contributors, sponsorships, or community channels like Discord/Slack. The project structure and use of modern development tools (FastAPI, Vue 3, Ruff, pytest) suggest active development.

Licensing & Compatibility

The project is released under the MIT license. This permissive license allows for broad compatibility, including commercial use and integration into closed-source applications without significant restrictions.

Limitations & Caveats

The SkillHub plugin marketplace is explicitly stated as "not yet open." The Android device automation feature requires specific setup, including USB debugging or network configuration, and is primarily demonstrated with the Hema app. Core functionality relies on configured LLM API keys, with a fallback to local Ollama models.