LobsterAI: NetEase’s Open‑Source 24/7 Personal Assistant Worth Exploring

NetEase Youdao has open‑sourced LobsterAI, a full‑scene AI personal assistant that runs 24/7, featuring a Cowork mode for local sandbox execution, remote IM triggers, built‑in skills and easy extensibility, with step‑by‑step installation and configuration guidance.

Ubiquitous Tech
Ubiquitous Tech
Ubiquitous Tech
LobsterAI: NetEase’s Open‑Source 24/7 Personal Assistant Worth Exploring

NetEase Youdao has open‑sourced LobsterAI, a full‑scene personal‑assistant agent that operates 24 hours a day. The project’s homepage (https://lobsterai.youdao.com) provides the download package and source code.

The core of LobsterAI is the Cowork mode , built on the Claude Agent SDK. Unlike chat‑only bots, Cowork can invoke tools, manipulate files, and run command‑line operations in a local or sandboxed environment, autonomously planning tasks and producing multimodal results such as automatically generated data‑scraping scripts and analysis reports.

LobsterAI supports remote control through major IM platforms—including DingTalk, Feishu, Telegram and Discord—so users can trigger the agent from a phone or any messaging client, enabling continuous background work such as scheduled tasks, email handling or video generation.

All execution happens locally, guaranteeing that sensitive data never leaves the machine. The system enforces strict directory boundaries and tool‑call approvals to protect privacy.

Out‑of‑the‑box, LobsterAI ships with a rich set of built‑in Skills covering office, learning and creative scenarios. Developers can extend the assistant by adding custom Skills; the article demonstrates adding a personal skill repository (https://github.com/wuchubuzai2018/expert‑skills‑hub) and shows the UI listing the newly added Skills.

Installation is straightforward: download the installer, run it, then configure the IM application ID and token (the article uses Feishu as an example). After that, configure the large‑model endpoint, add the skill repository URL, and the interface displays the available Skills.

A live demonstration shows the agent loading a “Winter Olympics” Skill, executing it, and returning the latest results. The same workflow is reproduced via a Feishu message, confirming that remote triggers and scheduled tasks work as described.

The source code is fully available on GitHub, allowing developers to study the architecture, avoid building an agent framework from scratch, and integrate LobsterAI’s sandboxed execution, Cowork mode and permission model into their own workflows.

Overall, LobsterAI provides a ready‑to‑use, locally secure, and extensible AI assistant that is valuable both for end users seeking a 24/7 digital helper and for developers looking to explore or customize an open‑source agent platform.

git clone
Original Source

Signed-in readers can open the original source through BestHub's protected redirect.

Sign in to view source
Republication Notice

This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactadmin@besthub.devand we will review it promptly.

Open SourceAI AgentPersonal AssistantCowork ModeLobsterAISkill Extension
Ubiquitous Tech
Written by

Ubiquitous Tech

A ubiquitous public account for pirate enthusiasts, regularly sharing curated experiences, tech learning, and growth insights. Currently publishing articles on AI RAG customer service, AI MCP technology, and open-source design. Personal free Knowledge Planet: Awakening New World Programmer.

0 followers
Reader feedback

How this landed with the community

Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.