Tbh I don’t know much about openclaw or moltbook, however given the nature of LLMs, it is probably not a good idea to let LLMs engage in such fashion.
As we speak, over 150 000 bots are networking, and I’m afraid we are only hours away from exploitation of this network by malicious actors.
It does not require a genius to recognise the enormous potential for a scale prompt injection.
These people are absolute idiots.