# Skills Review

> Review a skill for quality and framework compliance. Use when "review this skill", "check skill quality", "validate skill". Use when this capability is needed.

- Skill: `tomevault-io/skills-review` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add tomevault-io/skills-review`
- Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/skills-review/raw
- Safety review: pending (external: skill-scanner PASS, skillspector CAUTION)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: tomevault-io (https://skillmd.com/u/tomevault-io)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/tomevault-io/skills-review

---


Review one or more skills against the luca-skills framework (`framework/FRAMEWORK.md`).

### 1. Identify target

Determine which skill(s) to review.

- If the user specifies a skill by name, review that one.
- If no skill is specified, scan `skills/` and review all.
- Skip `skills-review` itself (don't self-review).

### 2. Check structure

For each skill folder, verify:

- `SKILL.md` exists (required — everything else is optional)
- YAML frontmatter is present and parseable
- Required frontmatter fields: `name`, `description`, `last-updated`
- File layout is reasonable:
  - Common optional files: `template.md`, `criteria.md`, `pending-updates.md`, `.skillstate.json`
  - Common machine-managed folders: `.backups/`
  - Extra files are allowed (some skills may include assets, examples, or helper scripts)

### 3. Check frontmatter

- `name` matches the directory name
- If `display-name` is present, it is plain language and suitable for showing in a list
- `description` is plain language and explains what the skill does
- `last-updated` is an ISO-8601 UTC timestamp (e.g. `2026-02-09T14:56:55Z`)
- If `inspirations` is present, each entry is a recognizable path (e.g. `owner/repo/path` or URL)
- If `depends` is present, each listed skill exists in `skills/`

### 4. Check procedure quality

Read the SKILL.md body and assess:

- Uses numbered steps with descriptive headers
- Imperative tone ("Do X", not "You should X")
- User-facing language is plain and non-technical (per `framework/REQUIREMENTS.md` § Interaction)
- References external files by path rather than embedding large blocks inline
- Covers the full workflow, not just the happy path

### 5. Check separation of concerns

This is a suggestion, not a hard rule. Flag if:

- Evaluation criteria are embedded in SKILL.md instead of living in `criteria.md`
- Output formatting is embedded in SKILL.md instead of living in `template.md`
- The skill is simple enough that separation would be over-engineering (note this and don't flag)

Reference: `framework/FRAMEWORK.md` § "Skill Folder Structure"

### 6. Report

Present findings grouped by severity:

**Errors** (must fix before the skill is usable):
- Missing SKILL.md
- Missing or unparseable frontmatter
- Missing required frontmatter fields (`name`, `description`, `last-updated`)
- Broken dependencies (listed in `depends` but not found)

**Warnings** (should fix):
- `name` doesn't match directory name
- `last-updated` isn't a valid ISO-8601 UTC timestamp
- Technical jargon in user-facing language
- Procedure lacks numbered steps or uses passive tone

**Suggestions** (consider):
- Separation of concerns could be improved
- Optional files (template.md, criteria.md) might benefit the skill
- Description could include more trigger phrases

If no issues are found, say so: "Skill passes all checks."

---
> Converted and distributed by [TomeVault](https://tomevault.io/claim/lucadellanna) — claim your Tome and manage your conversions.
<!-- tomevault:4.0:skill_md:2026-04-15 -->

