Open Claw cannot be made "secure" by any normal definition of the word. Unless I'm very much mistaken, fundamentally it's a tool that lets LLMs do stuff.

So you take the output of an LLM, which is obviously impossibly to guarantee correct, and use that to choose a tool and execute it. Like, send an email or whatever. And you take the input for that LLM not only from prompts, and various files, but also your system and random stuff you download from the internet.

I am telling you people, this is lunacy. No good can come of this.