liliMozi/openhanako

A personal AI agent with memory, personality, and autonomy.

What it solves

HanaAgent is designed to bridge the gap between powerful AI agents and non-technical users. It transforms the AI agent experience from a command-line interface into a user-friendly desktop application that can handle daily office tasks, manage files, and act as a personalized assistant with a distinct personality.

How it works

HanaAgent operates as a server-client architecture using an Electron-based desktop app and a Node.js server. It leverages the LLM providers (OpenAI, Anthropic, Ollama, etc.) and the Pi SDK for agent runtime. The system includes a memory system for long-term context, a personality template system for distinct agent behaviors, and a secure sandbox (using PathGuard and OS-level sandboxing like Seatbelt or Bubblewrap) to allow agents to safely execute code and interact with the system.

Who it’s for

It is built for everyone from office workers to developers, providing a GUI-based agent that can operate a computer, browse the web, and manage schedules without requiring complex configurations or technical jargon.

Highlights

  • Multi-Agent Collaboration: Create multiple agents with independent memories and personalities who can collaborate via group chats or delegate tasks to one another.
  • Coded Skills & Plugins: Supports a community ecosystem of skills that agents can install from GitHub or even write and learn new skills themselves.
  • Integrated Workspace: Each agent has a "desk" for asynchronous collaboration, where files and notes can be shared and monitored by the agent.
  • Cross-Platform Access: Connects to external platforms like Telegram, Feishu, QQ, and WeChat, allowing remote computer operation via these bots.
  • Security Sandbox: Implements a dual-layer isolation model to control agent permissions for system file access and network activity.

Related

  • Project
    zhayujie/CowAgent

    CowAgent is an open‑source, self‑hosted AI assistant that plans tasks, runs tools, keeps a personal knowledge base, and can operate as a team of cooperating agents. It supports many LLM providers, multimodal input, dozens of chat channels (Telegram, Slack, WeChat, etc.), a plug‑in skill system, and a three‑tier memory with nightly “Deep Dream” distillation. Install via a one‑line script or Docker, manage everything through a web console, and extend the system through the public Skill Hub or custom tool integrations.

  • Project
    the-open-agent/openagent

    An open-source personal AI assistant platform that combines LLMs, RAG, and autonomous agent loops into a single self-hostable binary.

  • Project
    theexperiencecompany/gaia

    An open-source personal AI assistant that automates daily administrative tasks across Gmail, Calendar, and other productivity tools via a multi-platform interface.

  • Project
    Stack-Cairn/LiveAgent

    A local-first AI agent desktop client that integrates LLM reasoning with local system tools for file manipulation, command execution, and remote browser-based control.

  • Project
    holaboss-ai/holaOS

    An agentic workspace that integrates AI agents side-by-side with real applications and chat tools, featuring local-first memory and shared tool access.