netease-youdao/LobsterAI

Open-source, desktop-grade AI agent that gets real work done — data analysis, slides, docs, video & web research. Built on OpenClaw; runs tools on your real desktop and takes commands from your phone via WeChat, Feishu, DingTalk & Telegram.

What it solves

LobsterAI is a desktop-grade AI agent designed to operate within a user's real working environment. It eliminates the gap between AI chat and actual execution by allowing an agent to interact with local files, terminal commands, browser workflows, and various office software (like Word, Excel, and PowerPoint) directly on the user's machine.

How it works

The project is split into two primary layers:

  • Cowork (Session Layer): A desktop application built with Electron and React that handles the user interface, local persistence (SQLite), permissions, and session history.
  • OpenClaw (Runtime Layer): The underlying engine and gateway that executes the agent's logic and tool calls.

It utilizes a system of "Skills" (built-in capabilities) and "Expert Kits" (curated workflow packages) to perform tasks. It also supports the Model Context Protocol (MCP) to connect to external data sources and tools, and integrates with various IM platforms (like WeChat, Telegram, and Discord) to allow remote control of the desktop agent.

Who it’s for

It is intended for professionals and power users who want to automate complex, multi-step office workflows, manage local project files, and run scheduled AI tasks directly on their desktop.

Highlights

  • Multi-Agent Workflows: Create specialized agents with unique identities, model choices, and skill sets.
  • IM Remote Control: Control your desktop agent remotely via platforms like WeChat, DingTalk, and Discord.
  • Harnessing Local Data: Direct access to local files, terminals, and workspace memory (e.g., MEMORY.md).
  • Scheduled Tasks: Ability to create recurring AI work, such as daily news digests or website monitoring.
  • Broad Tool Integration: 28 built-in skills covering everything from PDF processing to Remotion video generation.
  • MCP Support: Integration with Model Context Protocol servers for extended tool capabilities.

Related

  • Project
    limecloud/lime

    An open-source full-stack AI agent for Desktop and CLI that can edit code, run terminal commands, and manage multimodal research and content workflows.

  • Project
    hAcKlyc/MyAgents

    MyAgents is an open‑source, local‑first desktop workbench for creating and orchestrating AI agents. It supports multiple LLM runtimes, a task scheduler, IM channel integration, and a private device‑to‑device network (AgentNet) for multi‑computer workflows, all under an AGPL‑v3 license.

  • Project
    ValueCell-ai/ClawX

    ClawX is a desktop interface for OpenClaw AI agents that provides a graphical, no-terminal experience for managing AI workflows, scheduled tasks, and local computer automation.

  • 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
    clawdotnet/openclaw.net

    OpenClaw.NET is a .NET‑native, self‑hosted AI‑agent runtime with a desktop Avalonia UI (AgentQi Companion). It runs a gateway that talks to any OpenAI‑compatible model, offers streaming, tool execution, session memory, and many integrations (REST, WebSocket, MCP, 80+ tools, channel adapters). The gateway can be compiled with NativeAOT, making it suitable for edge or on‑prem deployments. Users can start via a ready‑made desktop bundle or from source, configure the assistant through chat commands, and extend functionality with .NET plugins or reusable MetaSkill DAG workflows.