How do I install the Review skill?

Run npx skillmds add majiayu000/review-48 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Review safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Review?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Review free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Review?

majiayu000 (@majiayu000) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Review by majiayu000 — Agent Skill

Name: Review
Rating: 4.2 (5 reviews)
Author: majiayu000

Review

(ePost) Use when user says "review", "check my code", "is this good", "look at this before I commit", or "suggest improvements" — detects review type (code quality, a11y, or general improvements) and runs the appropriate review

majiayu000 41e7fb5 2 files · 3.1 KB Updated Sep 9, 2026 567 repo stars

File contents

Review — Unified Review Command

Auto-detect and execute the appropriate review workflow.

Step 0 — Flag Override

If $ARGUMENTS starts with --code: load references/code.md and execute. If $ARGUMENTS starts with --a11y: load references/a11y.md and execute. If $ARGUMENTS starts with --ui: load references/ui-mode.md and execute. Delegate to epost-muji. If $ARGUMENTS starts with --improvements: run improvements inline (see below). Otherwise: continue to Auto-Detection.

Aspect Files

File	Purpose
`references/code.md`	Ultrathink edge cases, then parallel verify with reviewers
`references/a11y.md`	Review accessibility compliance (WCAG 2.1 AA)
`references/ui-mode.md`	Lightweight UI component review by focus area
`references/improvements.md`	Review auto-improvement metrics, detect patterns

Auto-Detection

Analyze $ARGUMENTS keywords:

Keyword match	Load Reference
"a11y", "accessibility", "wcag"	`references/a11y.md`
"ui", "component", "token", "klara", "muji"	`references/ui-mode.md` → epost-muji
"improvements", "metrics", "patterns"	Run improvements inline (see below)
Default (no keyword match)	`references/code.md`

Review-Improvements (Inline)

When dispatching review-improvements, run inline instead of forking (uses haiku model, restricted tools):

Read session metrics from .epost-data/improvements/sessions.jsonl
Read .epost-data/improvements/sessions.jsonl
Present findings grouped by severity (high → medium → low)
For each finding: explain detection, recommended action, next step
If no findings: report healthy, show session count
Summary table: severity × count

Execution

For code and a11y reviews: load the reference file and execute its workflow. For improvements: execute inline per the Review-Improvements section above.

---name: review-48description: (ePost) Use when user says "review", "check my code", "is this good", "look at this before I commit", or "suggest improvements" — detects review type (code quality, a11y, or general improvements) and runs the appropriate review--- # Review — Unified Review Command Auto-detect and execute the appropriate review workflow. ## Step 0 — Flag Override If `$ARGUMENTS` starts with `--code`: load `references/code.md` and execute. If `$ARGUMENTS` starts with `--a11y`: load `references/a11y.md` and execute. If `$ARGUMENTS` starts with `--ui`: load `references/ui-mode.md` and execute. Delegate to epost-muji. If `$ARGUMENTS` starts with `--improvements`: run improvements inline (see below). Otherwise: continue to Auto-Detection. ## Aspect Files | File | Purpose | |------|---------| | `references/code.md` | Ultrathink edge cases, then parallel verify with reviewers | | `references/a11y.md` | Review accessibility compliance (WCAG 2.1 AA) | | `references/ui-mode.md` | Lightweight UI component review by focus area | Loads review/references/ui-mode.md | epost-muji | | `references/improvements.md` | Review auto-improvement metrics, detect patterns | ## Auto-Detection Analyze `$ARGUMENTS` keywords: | Keyword match | Load Reference | |--------------|----------------| | "a11y", "accessibility", "wcag" | `references/a11y.md` | | "ui", "component", "token", "klara", "muji" | `references/ui-mode.md` → epost-muji | | "improvements", "metrics", "patterns" | Run improvements inline (see below) | | Default (no keyword match) | `references/code.md` | ## Review-Improvements (Inline) When dispatching review-improvements, run inline instead of forking (uses haiku model, restricted tools): 1. Read session metrics from `.epost-data/improvements/sessions.jsonl` 2. Read `.epost-data/improvements/sessions.jsonl` 3. Present findings grouped by severity (high → medium → low) 4. For each finding: explain detection, recommended action, next step 5. If no findings: report healthy, show session count 6. Summary table: severity × count ## Execution For code and a11y reviews: load the reference file and execute its workflow. For improvements: execute inline per the Review-Improvements section above.

Frequently asked questions