What problem does it solve? As your collection of Claude skills grows, the same principles get repeated across many skills while your central rules files fall out of date. This Skill scans all installed skills, finds principles that appear in two or more of them, and proposes concrete additions or revisions to your rules files — keeping rules complete without manual cross-reading. ## Core Features & Use Cases - Deterministic Inventory: Shell scripts scan global and project-level skills plus existing rule files, producing structured JSON inventories with headings and metadata. - LLM Cross-Read Analysis: Batched subagents compare skill principles against the full rules text and assign verdicts — Append, Revise, New Section, New File, Already Covered, or Too Specific. - User-Controlled Application: Every candidate is presented with evidence, violation risk, and draft text; nothing is written to rules without explicit user approval. - Use Case: After installing a batch of new skills, run a monthly rules maintenance pass to surface recurring principles (e.g., LLM output validation) and promote them into your shared rules with full traceability saved to results.json. ## Quick Start Ask Claude to run the rules-distill skill to scan my installed skills and propose updates to my rules files.
