Original Summary

Hi Reddit, TL;DR: While looking for a harness to use at my company, I built the engine that I'm open-sourcing today under AGPL-3.0. You can create AI assistants with long-term memory, many connectors (Google, Slack, Notion, etc) and a sandbox to run commands, all on your own server and with the models you choose. A single installation can serve a family, a company or a community, with each user isolated in their own account. The engine also works for assistants that serve the public over WhatsApp or other messaging interfaces. This is my first open source project, and anyone who wants to try it, contribute, or just talk about it is very welcome. The whole story: I'm the technical founder of a startup that builds virtual try-on technology for makeup, skin analysis models and beauty product recommendations. Last year we started using AI assistants to create hyper-personalized journeys in beauty retail. This year, especially after the Open Claw craze in January/February, I (like many other founders) went crazy over the impact agents could have on my company. I started looking for a harness that both our clients and our team could use safely. After experimenting with frameworks like Agno and LangChain to build something for my clients, and then with a few flavors of Claws to bring agents to my team, I decided to build a proprietary solution from scratch. I needed something that would bring assistants into the day-to-day of my operation, across every area of the business, supporting the clients, partners facing agents and, internally, the ones for my small team. The first test version came out in August. I published it as a closed beta for friends and family. Some people asked if they could bring more friends into the beta. I opened it up in September because I felt the platform was mature enough to take on some more people. Today the site is live in open beta, with a small group of active users. I also managed to make the system secure enough to serve what our enterprise clients need, our agentic business solutions are now being built over this engine. But what does this have to do with open-sourcing it? I expected this kind of platform to play a big role in society. I was at NVIDIA GTC and heard Jensen say that CEOs should have a claw strategy. I understood that it could change how people go through many parts of their lives (including retail, which interests me in particular because of my startup). What I didn't expect was a movement this big, with so much money in such a short time and such a huge amount of users handing over their data. I watched the funding announcements: Town.com (35M), then Instinct (350M), then Instinct again (1B), while I was here running my humble tests. Watching this huge movement, I thought about the amount of data and the power these companies will have by centralizing so much information. Today users tend to trade any data for convenience (I can say that because I'm seeing it with the service I put onlin

  • 情报分类：商业与市场研究
  • 分类依据：内容涉及商业、投资或市场动态
  • 信息来源：Reddit · SideProject
  • 发布时间：2026/10/17 20:45:41