Clawdbot AI assistant: What it is, how to try it

Updated on Tuesday, Jan. 27 at 5:11 p.m. ET — Clawdbot has officially changed its name, for very predictable reasons. Creator Peter Steinberger wrote on X that he's changed the AI tool's name under pressure from Anthropic, which makes the Claude family of large-language models. Moving forward, Clawdbot will henceforth be known as Moltbot.

This Tweet is currently unavailable. It might be loading or has been removed.

Interest in Clawdbot, an open-source AI personal assistant, has been building from a simmer to a roar. Over the weekend, online chatter about the tool reached viral status — at least, as viral as an open-source AI tool can be.

Clawdbot has developed a cult following in the early adopter community, and AI nerds in Silicon Valley are obsessively sharing best practices and showing off their DIY Clawdbot setups. The free, open-source AI assistant is commonly run on a dedicated Mac Mini (though other setups are possible), with users giving it access to their ChatGPT or Claude accounts, as well as email, calendars, and messaging apps.

Clawdbot has gone so viral on X that it's reached meme status, with developers and fans sharing tongue-in-cheek memes about their Clawdbot setups.

This Tweet is currently unavailable. It might be loading or has been removed.

This Tweet is currently unavailable. It might be loading or has been removed.

This Tweet is currently unavailable. It might be loading or has been removed.

So, what is Clawdbot 🦞, how can you try it, and why is it suddenly the talk of the town in Silicon Valley?

Clawdbot is an AI personal assistant

As previously mentioned, Clawdbot is an open-source AI assistant that runs locally on your device. The tool was built by developer and entrepreneur Peter Steinberger, best known for creating and selling PSPDFKit. The tool is often associated with the lobster emoji, for reasons that should be obvious.

Clawdbot is an impressive example of agentic AI, meaning it's a tool that can act autonomously and complete multi-step actions on behalf of the user. The year 2025 was supposed to be the year of AI agents; instead, many high-profile agentic AI implementations failed to deliver results, and there's a growing sense that AI agents are hitting a wall.

However, Clawdbot users say that the tool delivers where previous assistants have failed. The personal AI assistant remembers everything you've ever told it, and users can also grant it access to their email, calendar, and docs. On top of that, Clawdbot can proactively take personalized action. So, not only does Clawdbot check your email, but it can send you a message the moment a high-priority email arrives.

Based on its viral success, I'd be shocked if Steinberger isn't being courted by AI companies like OpenAI and Anthropic. Mashable reached out to Steinberger to ask about Clawdbot, and we'll update this post if we receive a response.

How to try Clawdbot

Steinberger has uploaded the source code for Clawdbot to GitHub, and you can download, install, and start experimenting with Clawdbot right away. (Find Clawdbot on Github.)

That said, downloading and setting up Clawdbot isn't as simple as downloading a typical app or piece of software. You'll need some technical know-how to get Clawdbot running on your device. There are also some serious security and privacy concerns to consider. More on that in a moment.

You can run Clawdbot on Mac, Windows, and Linux devices, and the Clawdbot website has installation instructions, system requirements, and tips.

Don't try Clawdbot without understanding the risks

Part of the reason that Clawdbot succeeds where other AI agents have failed is that it has full system access to your device. That means it can read and write files, run commands, execute scripts, and control your browser.

Steinberger is clear about the fact that running Clawdbot carries certain risks.

"Running an AI agent with shell access on your machine is… spicy," an FAQ reads. "Clawdbot is both a product and an experiment: you’re wiring frontier-model behavior into real messaging surfaces and real tools. There is no 'perfectly secure' setup." (Emphasis in original.)

Users can access a security audit tool for Clawdbot on GitHub, and the Clawdbot FAQ also has a useful security section. A sub-section titled "The Threat Model" notes that bad actors could "Try to trick your AI into doing bad things" and "Social engineer access to your data."

Frequently Asked Questions