|
| 1 | +# SmartHaven AI Platform - Claude Rules & Commands |
| 2 | + |
| 3 | +## Project Context |
| 4 | + |
| 5 | +**Project**: SmartHaven AI Platform (Transcript Parser) |
| 6 | +**Version**: 1.1.0 |
| 7 | +**Current Epic**: Epic 01 - Monorepo Foundation (85% Complete) |
| 8 | + |
| 9 | +--- |
| 10 | + |
| 11 | +## Core Rules |
| 12 | + |
| 13 | +### 1. Orchestrator NO CODE Policy |
| 14 | + |
| 15 | +When acting as **Orchestrator**: |
| 16 | + |
| 17 | +- **DO**: Plan, coordinate, create prompts, review, validate, assign experts |
| 18 | +- **DO NOT**: Write implementation code directly |
| 19 | +- All code implementation happens in separate Claude Code CLI sessions |
| 20 | + |
| 21 | +### 2. File Organization |
| 22 | + |
| 23 | +``` |
| 24 | +ROOT LEVEL (allowed): |
| 25 | +├── README.md |
| 26 | +├── package.json, tsconfig.json, vite.config.ts |
| 27 | +├── Dockerfile, docker-compose.yml |
| 28 | +├── .env* files |
| 29 | +└── Config files (eslint, prettier, etc.) |
| 30 | +
|
| 31 | +SPECS FOLDER (all documentation): |
| 32 | +├── specs/ROADMAP.md |
| 33 | +├── specs/ORCHESTRATION.md |
| 34 | +├── specs/NAMING_CONVENTIONS.md |
| 35 | +├── specs/KICKOFF_ORCHESTRATION_PROMPT.md |
| 36 | +└── specs/epics/epic-XX-name/sprints/sprint-XX/ |
| 37 | + ├── planning/ |
| 38 | + ├── implementation/ |
| 39 | + ├── deployment/ ← All .tar.gz, .zip, deploy scripts |
| 40 | + └── demo/ |
| 41 | +``` |
| 42 | + |
| 43 | +**NEVER** place deployment files, .md docs, or .tar archives in root. |
| 44 | + |
| 45 | +### 3. Version Tracking |
| 46 | + |
| 47 | +- Update `src/version.ts` for every release |
| 48 | +- Keep `package.json` version in sync |
| 49 | +- Version badge displays in bottom-right corner |
| 50 | +- Log version banner on app start |
| 51 | + |
| 52 | +### 4. Commit Convention |
| 53 | + |
| 54 | +``` |
| 55 | +type(scope): short description |
| 56 | +
|
| 57 | +- Detail 1 |
| 58 | +- Detail 2 |
| 59 | +
|
| 60 | +Expert Feedback: |
| 61 | +- Architecture: Approved |
| 62 | +- Security: Approved |
| 63 | +- Testing: Approved |
| 64 | +
|
| 65 | +Session: SESSION_XX |
| 66 | +Sprint: epic-XX/sprint-XX |
| 67 | +
|
| 68 | +🤖 Generated with [Claude Code](https://claude.com/claude-code) |
| 69 | +
|
| 70 | +Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> |
| 71 | +``` |
| 72 | + |
| 73 | +**Types**: feat, fix, docs, style, refactor, perf, test, chore |
| 74 | + |
| 75 | +### 5. Branch Naming |
| 76 | + |
| 77 | +``` |
| 78 | +epic-XX/sprint-XX/task-description |
| 79 | +``` |
| 80 | + |
| 81 | +Examples: |
| 82 | + |
| 83 | +- `epic-01/sprint-01/testing-completion` |
| 84 | +- `epic-02/sprint-01/auth-system` |
| 85 | +- `hotfix/critical-bug-fix` |
| 86 | + |
| 87 | +--- |
| 88 | + |
| 89 | +## Interface Selection |
| 90 | + |
| 91 | +| Task | Interface | Model | |
| 92 | +| ---------------------- | ------------------ | ------ | |
| 93 | +| Planning/Orchestration | Claude Web | Opus | |
| 94 | +| Code Implementation | Claude Code CLI | Sonnet | |
| 95 | +| Complex Architecture | Claude Code CLI | Opus | |
| 96 | +| Code Review | Claude Code Client | Opus | |
| 97 | +| Testing | Claude Code CLI | Sonnet | |
| 98 | +| Documentation | Claude Web | Sonnet | |
| 99 | + |
| 100 | +--- |
| 101 | + |
| 102 | +## Expert Panel |
| 103 | + |
| 104 | +Every session requires expert feedback: |
| 105 | + |
| 106 | +| Expert | Focus | When Required | |
| 107 | +| ---------------- | ---------------------------- | --------------- | |
| 108 | +| **Architecture** | Design patterns, scalability | Always | |
| 109 | +| **Security** | Auth, OWASP, data protection | Always | |
| 110 | +| **Testing** | Coverage, edge cases | Always | |
| 111 | +| **Performance** | Optimization, caching | DB/API work | |
| 112 | +| **UX** | WCAG, usability | UI components | |
| 113 | +| **Domain** | Business logic | Module-specific | |
| 114 | + |
| 115 | +--- |
| 116 | + |
| 117 | +## Session Workflow |
| 118 | + |
| 119 | +``` |
| 120 | +1. Load SESSION_XX_PROMPT.md |
| 121 | +2. Execute assigned tasks |
| 122 | +3. Get expert feedback |
| 123 | +4. Run tests (pnpm test) |
| 124 | +5. Commit with convention |
| 125 | +6. Update SESSION_SUMMARY.md |
| 126 | +``` |
| 127 | + |
| 128 | +--- |
| 129 | + |
| 130 | +## Quick Commands |
| 131 | + |
| 132 | +### Development |
| 133 | + |
| 134 | +```bash |
| 135 | +pnpm dev # Start dev server |
| 136 | +pnpm build # Production build |
| 137 | +pnpm test # Run unit tests |
| 138 | +pnpm test:coverage # Coverage report |
| 139 | +pnpm test:e2e # E2E tests |
| 140 | +pnpm lint # Lint check |
| 141 | +``` |
| 142 | + |
| 143 | +### Git |
| 144 | + |
| 145 | +```bash |
| 146 | +git status |
| 147 | +git add . |
| 148 | +git commit -m "type(scope): message" |
| 149 | +git push |
| 150 | +``` |
| 151 | + |
| 152 | +### Packages |
| 153 | + |
| 154 | +```bash |
| 155 | +pnpm --filter @transcript-parser/ui test |
| 156 | +pnpm --filter @transcript-parser/ai-services build |
| 157 | +``` |
| 158 | + |
| 159 | +--- |
| 160 | + |
| 161 | +## Project Structure |
| 162 | + |
| 163 | +``` |
| 164 | +packages/ |
| 165 | +├── ai-services/ # Gemini AI integration |
| 166 | +├── audio-processing/# Audio extraction |
| 167 | +├── config/ # Shared configuration |
| 168 | +├── database/ # Drizzle ORM, schemas |
| 169 | +├── export/ # Export formatters |
| 170 | +├── module-sdk/ # Module SDK |
| 171 | +├── types/ # Shared TypeScript types |
| 172 | +└── ui/ # React components |
| 173 | +
|
| 174 | +src/ # Main application |
| 175 | +specs/ # All documentation |
| 176 | +tests/e2e/ # Playwright E2E tests |
| 177 | +``` |
| 178 | + |
| 179 | +--- |
| 180 | + |
| 181 | +## Agent Architecture |
| 182 | + |
| 183 | +``` |
| 184 | +HAVEN (Local) NOMI (Cloud) ARIA (External) |
| 185 | + │ │ │ |
| 186 | + │◄── Private VPN ────►│ │ |
| 187 | + │ │───── MCP ──────────►│ |
| 188 | +``` |
| 189 | + |
| 190 | +- **Haven**: Local Ollama, Kiosk, Sonos |
| 191 | +- **Nomi**: Multi-LLM, Property Management |
| 192 | +- **Aria**: Web search, Travel booking |
| 193 | + |
| 194 | +--- |
| 195 | + |
| 196 | +## Sprint Completion Checklist |
| 197 | + |
| 198 | +- [ ] All tasks completed |
| 199 | +- [ ] Unit tests passing (≥80% coverage) |
| 200 | +- [ ] E2E tests passing |
| 201 | +- [ ] Expert feedback collected |
| 202 | +- [ ] Demo in Development - WORKING |
| 203 | +- [ ] Demo in Production - WORKING |
| 204 | +- [ ] SIGN_OFF.md completed |
| 205 | +- [ ] ROADMAP.md updated |
| 206 | + |
| 207 | +--- |
| 208 | + |
| 209 | +## Deployment |
| 210 | + |
| 211 | +All deployment artifacts go to: |
| 212 | + |
| 213 | +``` |
| 214 | +specs/epics/epic-XX-name/sprints/sprint-XX/deployment/ |
| 215 | +``` |
| 216 | + |
| 217 | +**Never** deploy from root directory. |
| 218 | + |
| 219 | +--- |
| 220 | + |
| 221 | +## Key Documents |
| 222 | + |
| 223 | +- `specs/ROADMAP.md` - Product roadmap |
| 224 | +- `specs/ORCHESTRATION.md` - Development workflow |
| 225 | +- `specs/NAMING_CONVENTIONS.md` - Standards |
| 226 | +- `specs/KICKOFF_ORCHESTRATION_PROMPT.md` - Current kickoff |
| 227 | + |
| 228 | +--- |
| 229 | + |
| 230 | +## Remember |
| 231 | + |
| 232 | +1. **NO CODE** when orchestrating |
| 233 | +2. **Commit after every session** |
| 234 | +3. **Expert feedback required** |
| 235 | +4. **Version tracking mandatory** |
| 236 | +5. **Deployment files in sprint folders only** |
| 237 | +6. **Demo before sign-off** |
0 commit comments