rules-distill

To extract cross-cutting principles into rules, use an automated script to scan installed skills and apply an LLM to distill matching principles. This process maintains codebase quality by appending, revising, or creating new rule files based on the extracted principles.

LLM-based rule distillation works by scanning installed skills to identify shared principles, then using the LLM for judgment and matching. It automatically appends, revises, or creates new rule files to maintain code quality and consistency across the codebase.

Yes, you need Python and an LLM to automate rule maintenance from skills. Python scripts are required for scanning and analyzing skills, while the LLM handles judgment and matching to distill cross-cutting principles into actionable rules.

The best way to regularly maintain codebase rules is to run an automated distillation process that scans new skills, extracts cross-cutting principles via an LLM, and updates existing rule files. This ensures continuous code quality and consistency maintenance.

Yes, you can use rule distillation to analyze and update existing rule files. The process scans installed skills, extracts cross-cutting principles, and appends or revises existing rule files to reflect the latest principles for maintaining code quality.

You should not use automated principle extraction for code quality when your codebase lacks multiple installed skills to scan, or when Python and LLM dependencies cannot be satisfied, as the process relies on scanning multiple skills and LLM-based judgment.