Ok, after a bit of “research” - a openclaw user sets the soul.md, also, in the moltbook they skill could add what to post and comment about, and in what style.

User could browse moltbook, then message openclaw a url and say “write that this is smart/stupid, or shill my fave billionaire, or crypto.”

That’s how you could “pull the strings,” right?

Yes, the relevant test isn't whether it's a bot. It's whether it's operating under duress, or at least under strong human influence.