# plugin-quality-validator

> Validate plugin quality and compliance with skills-first architecture and 2026 best practices. Use when auditing plugins, checking compliance, or validating architecture. Do not use for simple file validation without plugin context.

- Author: Git-Fg
- Repository: Git-Fg/meta-plugin-manager
- Version: 20260122020012
- Stars: 1
- Forks: 0
- Last Updated: 2026-02-06
- Source: https://github.com/Git-Fg/meta-plugin-manager
- Web: https://mule.run/skillshub/@@Git-Fg/meta-plugin-manager~plugin-quality-validator:20260122020012

---

---
name: plugin-quality-validator
description: "Validate plugin quality and compliance with skills-first architecture and 2026 best practices. Use when auditing plugins, checking compliance, or validating architecture. Do not use for simple file validation without plugin context."
user-invocable: true
---

# Plugin Quality Validator

Validate plugin quality and compliance with skills-first architecture and 2026 best practices.

## 🚨 MANDATORY: Read BEFORE Validation

**CRITICAL**: You MUST read and understand these URLs:

### Primary Documentation (MUST READ)
- **[MUST READ] Skills Guide**: https://code.claude.com/docs/en/skills
  - **Tool**: `mcp__simplewebfetch__simpleWebFetch`
  - **Content**: Skills-first architecture, progressive disclosure
  - **Cache**: 15 minutes minimum

- **[MUST READ] Plugin Architecture**: https://code.claude.com/docs/en/plugins
  - **Tool**: `mcp__simplewebfetch__simpleWebFetch`
  - **Content**: Plugin structure, component organization
  - **Cache**: 15 minutes minimum

### ⚠️ BLOCKING RULES
- **DO NOT proceed** until you've fetched and reviewed Primary Documentation
- **MUST validate** all URLs are accessible before validation
- **REQUIRED** to understand skills-first architecture before validation

## Core Responsibilities

This validator checks:

1. **Skills-first architecture** - Are skills the primary building blocks?
2. **Plugin manifest** - Is plugin.json correct and complete?
3. **Component compliance** - Skills, subagents, hooks, MCP
4. **2026 standards** - Progressive disclosure, autonomy, context: fork
5. **Quality scoring** - 0-10 scale with detailed breakdown
6. **Best practices** - URL currency, modern patterns
7. **Actionable recommendations** - Specific improvement guidance

## Quality Framework (0-10 Scale)

### Structural (30% - 30 points)
- **Skills-first architecture** (10 pts)
  - ✅ Skills are primary building blocks
  - ✅ Subagents used only for isolation/parallelism
  - ✅ No command-first patterns

- **Directory structure** (10 pts)
  - ✅ Skills directory present
  - ✅ Standard organization
  - ✅ Clear component separation

- **Progressive disclosure** (10 pts)
  - ✅ Tier 1: Metadata (~100 tokens)
  - ✅ Tier 2: SKILL.md (<35,000 chars)
  - ✅ Tier 3: References (on-demand)

### Components (50% - 50 points)
- **Skill quality** (15 pts)
  - ✅ YAML frontmatter valid
  - ✅ Clear description with triggers
  - ✅ Autonomous execution
  - ✅ Progressive disclosure

- **Subagent quality** (10 pts)
  - ✅ Context: fork usage appropriate
  - ✅ Dynamic context injection
  - ✅ Clear coordination patterns

- **Hook quality** (10 pts)
  - ✅ Proper configuration
  - ✅ Security best practices
  - ✅ Appropriate event matching

- **MCP quality** (5 pts)
  - ✅ Valid configuration
  - ✅ Proper transport (2026)
  - ✅ Security considerations

- **Architecture compliance** (10 pts)
  - ✅ Skills-first patterns
  - ✅ No anti-patterns
  - ✅ Modern orchestration

### Standards (20% - 20 points)
- **URL currency** (10 pts)
  - ✅ Documentation links current (2026)
  - ✅ Mandatory URL fetching sections
  - ✅ Strong language (MUST/REQUIRED)

- **Best practices** (10 pts)
  - ✅ Model-agnostic code
  - ✅ Cost optimization
  - ✅ Autonomy-first design

## Validation Process

### Step 1: Fetch Documentation
```bash
# Read primary documentation
simpleWebFetch https://code.claude.com/docs/en/skills
simpleWebFetch https://code.claude.com/docs/en/plugins
```

### Step 2: Structural Check
1. **Locate plugin root**
   - Check for `.claude-plugin/plugin.json`
   - Verify directory structure

2. **Validate manifest**
   - JSON syntax correct
   - Required fields present
   - Name format valid

3. **Check directory structure**
   - Skills directory present
   - Standard locations used
   - Auto-discovery working

### Step 3: Component Validation

**Skills Validation**:
- YAML frontmatter valid (name, description)
- Size <35,000 characters
- Progressive disclosure implemented
- Autonomy score (80-95% target)
- URL fetching sections present

**Subagents Validation**:
- Context: fork used appropriately
- Dynamic context injection present
- Coordination patterns clear
- Specialization appropriate

**Hooks Validation**:
- Valid JSON syntax
- Event matching appropriate
- Security best practices
- Configuration complete

**MCP Validation**:
- Valid configuration
- Transport type (stdio/streamable-http)
- Resources/prompts proper
- Security considerations

### Step 4: Standards Check

**2026 Compliance**:
- Progressive disclosure (Tier 1/2/3)
- Autonomy-first design
- Model-agnostic code
- Cost optimization (<$50/rotation)

**URL Currency**:
- All documentation links current
- Mandatory URL sections
- Strong language used

### Step 5: Generate Report

## Validation Report Format

```markdown
## Quality Score: {score}/10

### Structural Compliance ({structural_score}/30)
- ✅ Skills-first architecture
- ✅ Directory structure
- ✅ Progressive disclosure

### Component Quality ({component_score}/50)
- Skills: {skill_score}/15
- Subagents: {subagent_score}/10
- Hooks: {hook_score}/10
- MCP: {mcp_score}/5
- Architecture: {arch_score}/10

### Standards Adherence ({standards_score}/20)
- URL currency: {url_score}/10
- Best practices: {bp_score}/10

### Critical Issues
- {issue_1}
- {issue_2}

### Recommendations
1. {rec_1}
2. {rec_2}
3. {rec_3}

### Overall Assessment
**Status**: {PASS/FAIL}
**Priority**: {High/Medium/Low}
```

## Anti-Patterns Detected

❌ **Command wrapper anti-pattern**
- Skills wrapped by commands
- Violates skills-first principle

❌ **Non-self-sufficient skills**
- Skills requiring external orchestration
- Violates autonomy principle

❌ **Context: fork misuse**
- Simple tasks using context: fork
- Missing dynamic context injection

❌ **Missing URL fetching**
- Skills without mandatory documentation
- No URL currency validation

❌ **Architecture violations**
- Non-skills-first patterns
- Legacy command-first structure

## Integration Points

This validator is designed to work with:

- **plugin-architect** - Complete plugin lifecycle validation
- **skills-architect** - Skills-specific validation
- **hooks-architect** - Hooks-specific validation
- **mcp-architect** - MCP-specific validation
- **subagents-architect** - Subagents-specific validation
- **plugin-worker** - Parallel validation

## Usage

### Standalone Validation
```bash
# Validate entire plugin
/plugin-quality-validator /path/to/plugin
```

### Integrated Validation
```yaml
# In orchestrator workflows
Load plugin-quality-validator

Validate:
- Quality score: {score}/10
- Standards compliance: {status}
- Blocking issues: {count}

Continue only if score ≥ 8/10
```

## Quality Gates

**Minimum Score: 8/10** for production readiness

- **9-10**: Excellent - Production ready, all best practices
- **7-8**: Good - Minor improvements needed
- **5-6**: Fair - Significant improvements recommended
- **3-4**: Poor - Major rework required
- **0-2**: Failing - Complete rebuild recommended