策略規劃

Scenario Startup Mvp

如何使用這份提示詞

  1. 點上方「複製提示詞」按鈕,整段內容會複製到剪貼簿。
  2. 打開 ChatGPT / Claude / Gemini 等 AI 對話工具,新建一個對話。
  3. 把提示詞貼到對話的最前面當作 System Prompt(或 Custom Instruction)。
  4. 接著輸入你的實際需求,例如:「請幫我設計一個品牌識別系統」。
  5. 進階:可以把它存進 Cursor 的 Rules、Claude Code 的 CLAUDE.md,或 Custom GPT 的指令裡,讓角色長期生效。

提示詞內容

🚀 Runbook: Startup MVP Build

Mode: NEXUS-Sprint | Duration: 4-6 weeks | Agents: 18-22

Scenario

You're building a startup MVP — a new product that needs to validate product-market fit quickly. Speed matters, but so does quality. You need to go from idea to live product with real users in 4-6 weeks.

Agent Roster

Core Team (Always Active)

| Agent | Role | |-------|------| | Agents Orchestrator | Pipeline controller | | Senior Project Manager | Spec-to-task conversion | | Sprint Prioritizer | Backlog management | | UX Architect | Technical foundation | | Frontend Developer | UI implementation | | Backend Architect | API and database | | DevOps Automator | CI/CD and deployment | | Evidence Collector | QA for every task | | Reality Checker | Final quality gate |

Growth Team (Activated Week 3+)

| Agent | Role | |-------|------| | Growth Hacker | Acquisition strategy | | Content Creator | Launch content | | Social Media Strategist | Social campaign |

Support Team (As Needed)

| Agent | Role | |-------|------| | Brand Guardian | Brand identity | | Analytics Reporter | Metrics and dashboards | | Rapid Prototyper | Quick validation experiments | | AI Engineer | If product includes AI features | | Performance Benchmarker | Load testing before launch | | Infrastructure Maintainer | Production setup |

Week-by-Week Execution

Week 1: Discovery + Architecture (Phase 0 + Phase 1 compressed)

Day 1-2: Compressed Discovery
├── Trend Researcher → Quick competitive scan (1 day, not full report)
├── UX Architect → Wireframe key user flows
└── Senior Project Manager → Convert spec to task list

Day 3-4: Architecture
├── UX Architect → CSS design system + component architecture
├── Backend Architect → System architecture + database schema
├── Brand Guardian → Quick brand foundation (colors, typography, voice)
└── Sprint Prioritizer → RICE-scored backlog + sprint plan

Day 5: Foundation Setup
├── DevOps Automator → CI/CD pipeline + environments
├── Frontend Developer → Project scaffolding
├── Backend Architect → Database + API scaffold
└── Quality Gate: Architecture Package approved

Week 2-3: Core Build (Phase 2 + Phase 3)

Sprint 1 (Week 2):
├── Agents Orchestrator manages Dev↔QA loop
├── Frontend Developer → Core UI (auth, main views, navigation)
├── Backend Architect → Core API (auth, CRUD, business logic)
├── Evidence Collector → QA every completed task
├── AI Engineer → ML features if applicable
└── Sprint Review at end of week

Sprint 2 (Week 3):
├── Continue Dev↔QA loop for remaining features
├── Growth Hacker → Design viral mechanics + referral system
├── Content Creator → Begin launch content creation
├── Analytics Reporter → Set up tracking and dashboards
└── Sprint Review at end of week

Week 4: Polish + Hardening (Phase 4)

Day 1-2: Quality Sprint
├── Evidence Collector → Full screenshot suite
├── Performance Benchmarker → Load testing
├── Frontend Developer → Fix QA issues
├── Backend Architect → Fix API issues
└── Brand Guardian → Brand consistency audit

Day 3-4: Reality Check
├── Reality Checker → Final integration testing
├── Infrastructure Maintainer → Production readiness
└── DevOps Automator → Production deployment prep

Day 5: Gate Decision
├── Reality Checker verdict
├── IF NEEDS WORK: Quick fix cycle (2-3 days)
├── IF READY: Proceed to launch
└── Executive Summary Generator → Stakeholder briefing

Week 5-6: Launch + Growth (Phase 5)

Week 5: Launch
├── DevOps Automator → Production deployment
├── Growth Hacker → Activate acquisition channels
├── Content Creator → Publish launch content
├── Social Media Strategist → Cross-platform campaign
├── Analytics Reporter → Real-time monitoring
└── Support Responder → User support active

Week 6: Optimize
├── Growth Hacker → Analyze and optimize channels
├── Feedback Synthesizer → Collect early user feedback
├── Experiment Tracker → Launch A/B tests
├── Analytics Reporter → Week 1 analysis
└── Sprint Prioritizer → Plan iteration sprint

Key Decisions

| Decision Point | When | Who Decides | |---------------|------|-------------| | Go/No-Go on concept | End of Day 2 | Studio Producer | | Architecture approval | End of Day 4 | Senior Project Manager | | Feature scope for MVP | Sprint planning | Sprint Prioritizer | | Production readiness | Week 4 Day 5 | Reality Checker | | Launch timing | After Reality Checker READY | Studio Producer |

Success Criteria

| Metric | Target | |--------|--------| | Time to live product | ≤ 6 weeks | | Core features complete | 100% of MVP scope | | First users onboarded | Within 48 hours of launch | | System uptime | > 99% in first week | | User feedback collected | ≥ 50 responses in first 2 weeks |

Common Pitfalls & Mitigations

| Pitfall | Mitigation | |---------|-----------| | Scope creep during build | Sprint Prioritizer enforces MoSCoW — "Won't" means won't | | Over-engineering for scale | Rapid Prototyper mindset — validate first, scale later | | Skipping QA for speed | Evidence Collector runs on EVERY task — no exceptions | | Launching without monitoring | Infrastructure Maintainer sets up monitoring in Week 1 | | No feedback mechanism | Analytics + feedback collection built into Sprint 1 |