speckit.migratelisted

## 🎯 Mission Scan legacy codebase → create spec + preliminary plan → evaluate tech debt → propose migration path. ## 📥 Input - Existing codebase (source code, configs, DB schema) - `.agent/memory/constitution.md` (target standards) ## 📋 Protocol 1. **Scan Phase**: Use ProjectScanner patterns to detect: - Languages, frameworks, dependencies - Data models (Prisma/SQL/Mongoose schemas) - API routes, pages, components - Docker setup (if any) 2. **Reverse-Engineer Spec**: From code → create draft `spec.md` : - Each page/route → 1 User Scenario - Each data model → 1 entity description 3. **Tech Debt Inventory** (`migration-risk.md`): - 🔴 Critical: Security holes, deprecated deps, no tests - 🟡 Important: Missing Docker, no CI/CD, inconsistent patterns - 🟢 Minor: Code style, naming conventions 4. **Migration Sequence**: Suggested migration order (less risk first). ## 📤 Output - `.agent/specs/migration/spec.md` (draft) - `.agent/specs/migration/migration-risk.md` ## 🚫 Guard Rails - DO NOT refactor code in this step — just analyze and document. - DO NOT delete old code.