# Skill Library Analysis

> Analyze, deduplicate, cluster, and select skills from a large skill library or cloned upstream sources. Use when maintaining AI_Skills_Collection, comparing external skill repos, or deciding whether to merge, add, archive, or ignore skills.

- Skill: `yuukias/skill-library-analysis` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add yuukias/skill-library-analysis`
- Raw SKILL.md: https://api.skillmd.com/api/skills/yuukias/skill-library-analysis/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: YuukiAS (https://skillmd.com/u/yuukias)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/yuukias/skill-library-analysis

---

# Skill Library Analysis

Use this to keep a large skill collection coherent. The goal is to reduce duplicates and preserve provenance, not to maximize skill count.

## Workflow

1. Inventory candidate skills with path, name, description, provenance, license, and upstream commit.
2. Group by user task, not by repository.
3. Compare each group against existing local skills:
   - exact duplicate;
   - same trigger with stronger references;
   - complementary capability;
   - plugin-only or software-only source;
   - unclear license or provenance.
4. Choose one action:
   - merge into existing skill;
   - create one new local skill;
   - record only;
   - reject as software/non-skill;
   - defer because of conflict.
5. Record decisions in a tracked source log before deleting scratch clones.

## Quality Rules

- Do not bulk-import a repository when a distilled local skill is enough.
- Do not keep multiple active skills with the same trigger boundary.
- Do not hide conflicts; record them for user decision.
- Prefer profile and marketplace aggregation over many near-duplicate installable skills.