liliMozi/openhanako

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

What it solves

HanaAgent bridges the gap between complex AI agents and non-technical users by providing a user-friendly, graphical interface for a powerful AI assistant. It moves agent capabilities out of the command line and into a desktop application designed for daily office work and personal assistance, focusing on natural communication and intuitive interaction.

How it works

The system operates as a server-client architecture where a Node.js server (using Hono) manages the agent's core logic, memory, and tools, while an Electron-based desktop app provides the UI. It utilizes the Pi SDK for the agent runtime and supports multiple LLM providers (OpenAI, Anthropic, Ollama). The agent can interact with the local OS through a secure sandbox (PathGuard and OS-level sandboxing) to perform tasks like reading/writing files, browsing the web, and executing code.

Who it’s for

It is designed for anyone working on a computer, specifically targeting non-coders and office workers who want a personalized AI assistant that can manage schedules, handle files, and automate routine tasks without needing to write complex configurations.

Highlights

  • Personality & Memory: Supports custom personality templates and a memory system to ensure agents behave consistently and remember past interactions.
  • Multi-Agent Collaboration: Allows the creation of multiple agents with independent memories and personalities who can collaborate in group channels or delegate tasks.
  • Integrated Workspace: Each agent has a "desk" for files and notes, enabling asynchronous collaboration between the user and the AI.
  • Extensive Tooling: Capable of web browsing, internet searching, file manipulation, and executing terminal commands.
  • Skill System: Compatible with a broad skills ecosystem; agents can install community skills from GitHub or even write their own new skills.
  • Multi-Platform Access: Can be connected to Telegram, Lark, QQ, and WeChat, allowing users to remotely control their computer via these platforms.
  • Security Sandbox: Implements a dual-layer isolation model to restrict agent access to specific directories and system resources.

Related

  • 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
    zhayujie/CowAgent

    An open-source AI assistant framework that proactively plans tasks, manages long-term memory and knowledge, and controls computers across multiple messaging platforms.

  • 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.

  • Project
    AIDotNet/OpenCowork

    OpenCowork is an open‑source desktop (Electron + .NET) platform that lets LLM‑based agents work directly on your local files, shell, and repositories. It supports multi‑agent teams, integrates with many chat services, offers five interaction modes (chat, clarify, cowork, code, acp), and provides a rich toolbox (file/shell access, browser automation, task/cron management, CodeGraph indexing, skill extensions). The same native .NET worker powers both the GUI and a terminal‑based CLI, keeping a single authoritative agent loop. Quick start requires Node ≥ 18 and .NET 11 SDK; builds are available for Windows, macOS, and Linux. Use cases include autonomous coding, scheduled ops, research/report generation, remote SSH tasks, and media‑pipeline design. Licensed under Apache 2.0.