"Use compact prompt" feature from Cline for local / weaker models

[yuisheaven]

I would like to ask for the "Use compact prompt" feature that cline has already implemented. This is crucial for working with local LLMs because they get overwhelmed by the context of the currently very large system prompt.
Reference: https://cline.bot/blog/local-models