Tal
Tal
i am not sure i fully understand the issue, it if its indeed a `pr-agent` and not `litellm` issue But if you think specific treatment will help, you can suggest...
try again, added support https://github.com/qodo-ai/pr-agent?tab=readme-ov-file#aug-8-2025
gpt-5-mini works for me:
the auto answer above capture correctly the current abilities
pr agent currently uses spaces, not tabs. yes https://github.com/qodo-ai/pr-agent/blob/c635887949e7abbb6fd04a03cc76fa1d9b59d78a/pr_agent/tools/pr_code_suggestions.py#L589 https://github.com/qodo-ai/pr-agent/blob/c635887949e7abbb6fd04a03cc76fa1d9b59d78a/pr_agent/tools/pr_code_suggestions.py#L621 I would be open for a PR to also support tabs when a flag is on
Provide a fully reproducible example for how it breaks. "Shell script" is too general, and the error you described can be due to many reasons
The main time consumption is via LLM calls, and it can be logged with the existing litellm callback. Any other logging are more suitable for local implementation in your fork
Your command line, or your configuration, (which you have not shared) are wrong
how can 'github.user_token' be relevant for bitbucket ?? the example you are using is for github. you need to edit the script to be similar to this:  all the...