personalization

To personalize AI interactions across sessions, this Skill stores your stated preferences and dislikes in designated LLM context directories. It automatically records your natural language feedback to maintain a consistent user model.

To filter unwanted topics, simply tell the AI your dislikes in natural language. The preference tracking system applies personalized filtering to avoid those subjects in future interactions unless explicitly requested.

You can customize interaction style by stating your preference, such as requesting concise summaries. The AI adapts its communication style based on your direct feedback and stores this writing style preference for future sessions.

User preference tracking is the mechanism of recording stated likes and dislikes to tailor AI interactions. It works by storing self-model data, including personal profiles and writing styles, in LLM context directories.

No specific dependencies or frameworks are required to implement this personalization. It operates natively within conversational AI systems by managing user preferences and applying adaptive behavior based on stated boundaries.

The limitation of preference filtering is that it relies on explicitly stated boundaries. The AI will avoid unwanted topics unless explicitly asked, meaning dynamic or unmentioned subjects may still appear unless directly added to your dislikes.