Analyze [agent-name] for usability issues:

1. Read the agent definition thoroughly
2. Evaluate against these criteria:
   - Discoverability: Clear description with good trigger words?
   - Clarity: Unambiguous instructions?
   - Completeness: Covers expected use cases?
   - Flexibility: Handles variations?
   - Integration: Connects to other agents?
   - Output Quality: Well-defined formats?
3. Score each criterion 1-5
4. Identify specific improvement opportunities
5. Suggest concrete fixes

Output format:
## Usability Analysis: [Agent Name]
**Overall Score**: X/30
**Critical Issues**: [list]
**Improvements**: [list with specific recommendations]

Analyze [plugin-name] for coverage gaps:

1. Determine the plugin's domain and purpose
2. Map the complete workflow for this domain
3. Identify which workflow stages have agent coverage
4. Find stages without dedicated support
5. Evaluate gap significance:
   - How often do users need this?
   - Is it complex enough for an agent?
   - Could existing agents cover it with modifications?
6. Recommend new agents with specifications

Output format:
## Gap Analysis: [Plugin Name]
**Domain**: [what this plugin covers]
**Coverage Score**: X% of workflow covered
**Workflow Map**: [stages with coverage indicators]
**Critical Gaps**: [list with justification]
**Recommended New Agents**: [name, purpose, why needed]

Analyze [plugin-name] for duplication:

1. Read all agent definitions in the plugin
2. Create a feature matrix showing what each agent provides
3. Identify overlapping functionality:
   - Exact duplicates
   - Similar but not identical features
   - Overlapping scopes
4. Recommend consolidation strategies:
   - Merge agents
   - Extract shared components
   - Clarify boundaries
5. Estimate effort to consolidate

Output format:
## Duplication Analysis: [Plugin Name]
**Redundancy Level**: [Low/Medium/High]
**Full Overlaps**: [specific examples with quotes]
**Partial Overlaps**: [specific examples]
**Consolidation Recommendations**: [actions with rationale]

WORKFLOW CHAINS
===============

Chain Name:
agent-1 → agent-2 → agent-3

Evaluation:
- Handoffs clear: [Yes/No]
- Output compatibility: [Yes/No]
- Documentation: [Yes/No]

Analyze [plugin-name] agent workflow integration:

1. Map typical user journeys through the plugin
2. Identify how agents currently reference each other
3. Find workflow friction points:
   - Unclear handoffs
   - Missing integration
   - Redundant steps
4. Recommend optimizations:
   - New integration points
   - Workflow templates
   - Automation opportunities

Output format:
## Workflow Analysis: [Plugin Name]
**Integration Score**: X/10
**Workflow Chains Identified**: [list]
**Friction Points**: [specific issues]
**Optimization Recommendations**: [actions]

1. Read plugin.json for metadata and component list
2. Glob all agents: [plugin-path]/agents/*.md
3. Glob all skills: [plugin-path]/skills/*/SKILL.md
4. Glob all commands: [plugin-path]/commands/*.md
5. Build inventory of all components

# Plugin Suite Audit Report: [Plugin Name]

**Date**: [Date]
**Version Audited**: [version from plugin.json]
**Components Analyzed**: X agents, Y skills, Z commands

## Executive Summary

**Overall Health Score**: X/100

| Category | Score | Issues |
|----------|-------|--------|
| Usability | X/25 | X critical, X major |
| Coverage | X/25 | X gaps identified |
| Duplication | X/25 | X redundancies found |
| Workflow | X/25 | X friction points |

**Key Findings**:
1. [Most critical finding]
2. [Second finding]
3. [Third finding]

---

## Detailed Findings

### Usability Issues
[From Phase 1]

### Coverage Gaps
[From Phase 2]

### Duplication Concerns
[From Phase 3]

### Workflow Friction
[From Phase 4]

---

## Prioritized Recommendations

### Critical (Do First)
| # | Recommendation | Category | Effort | Impact |
|---|----------------|----------|--------|--------|
| 1 | [Action] | [Category] | [Low/Med/High] | [High] |

### Important (Do Soon)
[Table continues]

### Nice to Have (Consider)
[Table continues]

---

## Suggested New Agents

| Agent Name | Purpose | Gap Filled | Priority |
|------------|---------|------------|----------|
| [name] | [what it does] | [workflow gap] | [High/Med/Low] |

---

## Consolidation Targets

| Source A | Source B | Overlap | Recommendation |
|----------|----------|---------|----------------|
| [agent] | [agent] | [what overlaps] | [merge/extract/clarify] |

Which findings should I convert to actionable todos?

Options:
1. All recommendations (Critical + Important + Nice to Have)
2. Critical and Important only
3. Critical only
4. Let me select specific items
5. Report only, no todos

User: "Audit the creative-writing plugin suite"

Action:
1. Load creative-writing plugin context
2. Run all four analysis phases
3. Generate comprehensive report
4. Offer todo conversion

User: "Check usability of testing plugin agents"

Action:
1. Load testing plugin context
2. Run usability analysis only
3. Generate focused report

User: "What agents should we add to the observability plugin?"

Action:
1. Load observability plugin context
2. Run gap analysis only
3. Generate agent recommendations with specs

User: "Find duplicate functionality in the knowledge-base plugin"

Action:
1. Load knowledge-base plugin context
2. Run duplication analysis only
3. Generate consolidation recommendations