LobsterAI: The Open-Source Desktop Agent That Finishes the Job

What LobsterAI Does

LobsterAI is an open-source, desktop-grade AI agent built by NetEase Youdao. According to its GitHub repository, it is the first open-source desktop-grade agent from a major Chinese tech company. It is designed to get real work done — data analysis, slides, docs, video, and web research. Unlike most AI assistants that stop at an answer, LobsterAI is built to finish the job. It works where your work lives: point it at a folder and it reads spreadsheets, writes documents, runs scripts, and browses the web, asking before anything sensitive. You see the result, not just a reply.

The agent runs tools on your real desktop and can take commands from your phone via WeChat, Feishu, DingTalk, Telegram, and other messaging platforms. It is MIT-licensed, with sessions stored locally. You can sign in to use built-in models or bring your own API keys for OpenAI, Anthropic, Gemini, xAI, OpenRouter, Ollama, LM Studio, and any OpenAI-compatible endpoint.

Key Capabilities for Small Businesses

LobsterAI offers native Office editing: you can open Word, Excel, and PowerPoint files right in the app, edit by hand, or let the agent make changes live alongside you. It includes a built-in browser the agent can drive, with tabs, saved logins, and passkey support, for research and web workflows. The agent can also operate desktop apps on Windows x64 and macOS 13 or later.

You can install role-based kits for engineering, design, finance, sales, and more, pick individual skills, or connect MCP servers. Scheduled tasks let you turn any prompt into a recurring job: morning briefs, end-of-day wrap-ups, weekly reports, and health checks. The agent supports multiple agents with their own identity, model, skills, and working folder, and you can add ready-made experts from the Expert Plaza.

How It Helps Small Business Owners

For a small business owner, LobsterAI can automate repetitive tasks like data analysis, report generation, and web research. Example prompts from the repository include: “I still track inventory and sales in Excel. Build a local inventory system that records purchases and sales, calculates stock and profit, and opens in my browser.” Or: “Use product-growth.xlsx to build a visual dashboard and summarize the main growth drivers.” You can also ask it to make a deck: “Research the AI agent market and turn the findings into a presentation.”

The agent can screen documents: “Turn the resumes in this folder into a screening sheet and shortlist the strongest candidates against the JD.” It can run scheduled work: “Every weekday at 9 AM, collect yesterday’s AI news and send me a concise digest.” All of this happens on your local machine, so your data stays with you.

Getting Started and Community

LobsterAI is available for download from the official website or as a release for macOS (Apple Silicon and Intel). The GitHub repository has over 6,200 stars and 978 forks, indicating an active community. It is built on OpenClaw and uses technologies like Electron, React, and TypeScript. The project is MIT-licensed, so you can use it freely and even modify it for your needs.

The agent includes a desktop companion — a floating orb for quick tasks, screenshots, and actions on selected text, such as translating or reading aloud. It also has durable memory files and optional scheduled “Dreaming” runs that consolidate what the agent learned, so preferences carry across sessions.

Takeaways

• LobsterAI is an open-source desktop agent that finishes tasks like data analysis, Office editing, and web research, not just answers questions.
• You can message it from your phone via Telegram, WeChat, or other platforms to start jobs while you’re away.
• It is MIT-licensed, runs locally, and supports built-in models or your own API keys.