rules-distill

原文涵盖的主题

• Rules Distill
• When to Use
• How It Works
• Phase 1: Inventory (Deterministic Collection)
• 1a. Collect skill inventory
• 1b. Collect rules index
• 1c. Present to user
• Phase 2: Cross-read, Match & Verdict (LLM Judgment)

Scan installed skills, extract cross-cutting principles that appear in multiple skills, and distill them into rules — appending to existing rule files, revising outdated content, or creating new rule files. Applies the "deterministic collection + LLM judgment" principle: scripts collect facts exhaustively, then an LLM cross-reads the full context and produces verdicts. Periodic rules maintenance (monthly or after installing new skills) After a skill-stocktake reveals patterns that should be rules When rules feel incomplete relative to the skills being used The rules distillation process follows three phases: Extraction and matching are unified in a single pass. Rules files are small enough (800 lines total) that the full text can be provided to the LLM — no grep pre-filtering needed. Group skills into thematic clusters based on their descriptions. Analyze each cluster in a subagent with the full rules text. After all batches complete, merge candidates across batches: Deduplicate candidates with the same or overlapping principles Re-check the "2+ skills" requirement using evidence from all batches combined — a principle found in 1 skill per batch but 2+ skills total is valid Launch

这里只展示节选，请以完整原文和当前要求为准。