update ollama #6
Conversation
|
|
||
| ~~~bibtex | ||
| ┌──(venv)─(root㉿kali)-[/home/kali/Downloads/hackingBuddyGPT] | ||
| └─# cd src/hackingBuddyGPT/cli |
There was a problem hiding this comment.
hm, have you tried using the latest hackingBuddyGPT version (0.3.1)? You can install it with pip install hackingBuddyGPT and then can just call hackingBuddyGPT from the command line. This might make the setup easier!
| ┌──(venv)─(root㉿kali)-[/home/…/hackingBuddyGPT/src/hackingBuddyGPT/cli] | ||
| └─# python wintermute.py | ||
| usage: wintermute.py [-h] | ||
| {linux_privesc_hintfile,linux_privesc_guided,linux_privesc,windows_privesc,minimal_linux_privesc,minimal_linux_templated_agent,simple_web_test,web_test_with_explanation,simple_web_api_testing,simple_web_api_documentation} |
There was a problem hiding this comment.
those are named differently by now.. would it be possible for you to do the same with the latest version of hackingBuddyGPT? Functionally, nothing (or not much) should change but you should be able to directly call hackingBuddyGPT and the use-cases are named differently
There was a problem hiding this comment.
I will go back to comfirm and modify the file again, thank you for the reminder!!
|
|
||
| ~~~bibtex | ||
| //then copy the selected model into gpt-3.5-turbo | ||
| ollama cp gdisney/mistral-uncensored gpt-3.5-turbo |
There was a problem hiding this comment.
this will break some of the internal logic of hackingBuddyGPT (e.g., token estimation as this uses the given model name to switch between used tokenizers), maybe mention that this is not recommended?
2 participants
Added ollama settings, I will also add troubleshooting content in the future