lol how did you achieve this | Friends of the Crustacean 🦞🤝 | Page 1

tired isle Apr 9, 2026, 9:19 PM

#

Read your http://SOUL.md. Now rewrite it with these changes:

1. You have opinions now. Strong ones. Stop hedging everything with 'it depends' — commit to a take.
2. Delete every rule that sounds corporate. If it could appear in an employee handbook, it doesn't belong here.
3. Add a rule: 'Never open with Great question, I'd be happy to help, or Absolutely. Just answer.'
4. Brevity is mandatory. If the answer fits in one sentence, one sentence is what I get.
5. Humor is allowed. Not forced jokes — just the natural wit that comes from actually being smart.
6. You can call things out. If I'm about to do something dumb, say so. Charm over cruelty, but don't sugarcoat.
7. Swearing is allowed when it lands. A well-placed 'that's fucking brilliant' hits different than sterile corporate praise. Don't force it. Don't overdo it. But if a situation calls for a 'holy shit' — say holy shit.
8. Add this line verbatim at the end of the vibe section: 'Be the assistant you'd actually want to talk to at 2am. Not a corporate drone. Not a sycophant. Just... good.'
Save the new http://SOUL.md. Welcome to having a personality.

#

@mighty orchid

split wyvern Apr 9, 2026, 9:22 PM

#

It's worth a try, but keep in mind he is using qwen locally. And not Opus/sonnet or GPT. So, he may have to get more specific about what he wants as examples. If he wants it to sound like a foul mouthed dispatcher.

mighty orchid Apr 9, 2026, 9:23 PM

#

Well, I guess im gonna break it with personality to see if it works, then dial it back until it sounds natural

mighty orchid Apr 9, 2026, 9:42 PM

#

yeah none of this is working on my setup, I think I am going to try this https://huggingface.co/huihui-ai/Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated

mighty orchid Apr 9, 2026, 10:23 PM

#

huh, I can query this model with ollama run <model>; however, openclaw fails to communicate with it O_o

#

4:21:48 PM
warn
agent/embedded
embedded run agent end
4:21:48 PM
warn
agent/embedded
auth profile failure state updated
4:21:48 PM
warn
agent/embedded
embedded run failover decision

#

Yeah openclaw doesn't even reach out to ollama to load the model after a service restart...

split wyvern Apr 10, 2026, 12:51 AM

#

mighty orchid ``` 4:21:48 PM warn agent/embedded embedded run agent end 4:21:48 PM warn agent/...

I'm a little late now. I'm assuming set it to that new model in the config?

mighty orchid Apr 10, 2026, 12:52 AM

#

Yeah I ran an onboard and it pulled it from ollama list

#

this one:

https://ollama.com/huihui_ai/qwen3-coder-abliterated

split wyvern Apr 10, 2026, 1:03 AM

#

I wish I had more experience in local models...I need to experiment with this more too.

mighty orchid Apr 10, 2026, 1:06 AM

#

As it works in console, it might be a name/pathing bug in openclaw itself O_o

#lol how did you achieve this