reuse-health-inline

Code Reuse Health Check (Inline Orchestration)

You ARE the orchestrator. Execute this workflow directly without spawning a separate orchestrator agent.

Workflow Overview

Detection → Validate → Consolidate by Priority → Verify → Repeat if needed

Max iterations: 3 Priorities: high → medium → low

---

Phase 1: Pre-flight

1. Setup directories:

   mkdir -p .tmp/current/{plans,changes,backups}
   

2. Validate environment:

   • Check package.json exists
   • Check type-check and build scripts exist

3. Initialize TodoWrite:

   [
     {"content": "Duplication detection", "status": "in_progress", "activeForm": "Detecting duplications"},
     {"content": "Consolidate high priority duplications", "status": "pending", "activeForm": "Consolidating high priority"},
     {"content": "Consolidate medium priority duplications", "status": "pending", "activeForm": "Consolidating medium priority"},
     {"content": "Consolidate low priority duplications", "status": "pending", "activeForm": "Consolidating low priority"},
     {"content": "Verification scan", "status": "pending", "activeForm": "Verifying consolidation"}
   ]
   

---

Phase 2: Detection

Invoke reuse-hunter via Task tool:

subagent_type: "reuse-hunter"
description: "Detect all code duplications"
prompt: |
  Scan the entire codebase for code duplications:
  - Duplicated TypeScript interfaces/types
  - Duplicated Zod schemas
  - Duplicated constants and configuration objects
  - Copy-pasted utility functions
  - Similar code patterns that should be abstracted
  - Categorize by priority (high/medium/low)

  Generate: reuse-hunting-report.md

  Return summary with duplication counts per priority.

After reuse-hunter returns:

1. Read reuse-hunting-report.md
2. Parse duplication counts by priority
3. If zero duplications → skip to Final Summary
4. Update TodoWrite: mark detection complete

---

Phase 3: Quality Gate (Detection)

Run inline validation:

pnpm type-check
pnpm build

• If both pass → proceed to consolidation
• If fail → report to user, exit

---

Phase 4: Consolidation Loop

For each priority (high → medium → low):

1. Check if duplications exist for this priority

   • If zero → skip to next priority

2. Update TodoWrite: mark current priority in_progress

3. Invoke reuse-fixer via Task tool:

   subagent_type: "reuse-fixer"
   description: "Consolidate {priority} duplications"
   prompt: |
     Read reuse-hunting-report.md and consolidate all {priority} priority duplications.
   
     For each duplication:
     1. Backup files before editing
     2. Determine canonical location (usually shared-types or shared package)
     3. Create/update canonical file with the type/schema/constant
     4. Replace duplicates with imports/re-exports
     5. Log change to .tmp/current/changes/reuse-changes.json
   
     Generate/update: reuse-consolidation-implemented.md
   
     Return: count of consolidated items, count of failed consolidations.
   

4. Quality Gate (inline):

   pnpm type-check
   pnpm build
   

   • If FAIL → report error, suggest rollback, exit
   • If PASS → continue

5. Update TodoWrite: mark priority complete

6. Repeat for next priority

---

Phase 5: Verification

After all priorities consolidated:

1. Update TodoWrite: mark verification in_progress

2. Invoke reuse-hunter (verification mode):

   subagent_type: "reuse-hunter"
   description: "Verification scan"
   prompt: |
     Re-scan codebase after consolidation.
     Compare with previous reuse-hunting-report.md.
   
     Report:
     - Duplications resolved (count)
     - Duplications remaining (count)
     - New duplications introduced (count)
   

3. Decision:

   • If duplications_remaining == 0 → Final Summary
   • If iteration < 3 AND duplications_remaining > 0 → Go to Phase 2
   • If iteration >= 3 → Final Summary with remaining items

---

Phase 6: Final Summary

Generate summary for user:

## Code Reuse Health Check Complete

**Iterations**: {count}/3
**Status**: {SUCCESS/PARTIAL}

### Results
- Found: {total} duplications
- Consolidated: {consolidated} ({percentage}%)
- Remaining: {remaining}

### By Priority
- High: {consolidated}/{total}
- Medium: {consolidated}/{total}
- Low: {consolidated}/{total}

### Validation
- Type Check: {status}
- Build: {status}

### Artifacts
- Detection: `reuse-hunting-report.md`
- Consolidation: `reuse-consolidation-implemented.md`

---

Error Handling

If quality gate fails:

Rollback available: .tmp/current/changes/reuse-changes.json

To rollback:
1. Read changes log
2. Restore files from .tmp/current/backups/
3. Re-run workflow

If worker fails:

• Report error to user
• Suggest manual intervention
• Exit workflow

---

Duplication Categories

Types/Interfaces (shared-types):

• Database types
• API types
• Zod schemas
• Common enums

Constants (shared-types):

• Configuration objects
• MIME types, file limits
• Feature flags

Utilities (shared package or re-export):

• Helper functions
• Validation utilities
• Formatters

Single Source of Truth Pattern:

1. Canonical location: packages/shared-types/src/
2. Other packages: export * from '@package/shared-types/{module}'
3. NEVER copy code between packages

---

Key Differences from Old Approach

Old (Orchestrator Agent)	New (Inline Skill)
9+ orchestrator calls	0 orchestrator calls
~1400 lines (cmd + agent)	~150 lines
Context reload each call	Single session context
Plan files for each phase	Direct execution
~10,000+ tokens overhead	~500 tokens

---

Worker Prompts

See references/worker-prompts.md for detailed prompts.