Recommend technology stacks based on project requirements, team expertise, and constraints. Use when selecting frameworks, languages, databases, and infrastructure for new projects.
インストール方法を見るインストールする前に、エージェントに与えられる指示の中身を確認できます。
Provides structured recommendations for technology stack selection based on project requirements, team constraints, and business goals.
┌───────────────────────────────────────────────────────────────────┐
│ STACK SELECTION INPUTS │
├───────────────────────────────────────────────────────────────────┤
│ │
│ Project Requirements Team Factors Business Constraints│
│ ──────────────────── ──────────── ────────────────── │
│ • Scale expectations • Current skills • Time to market │
│ • Performance needs • Learning capacity • Budget │
│ • Integration points • Team size • Hiring market │
│ • Compliance/Security • Experience level • Long-term support │
│ │
└───────────────────────────────────────────────────────────────────┘
│
▼
┌─────────────────┐
│ RECOMMENDATION │
│ Framework │
└─────────────────┘
| Project Type | Frontend | Backend | Database | Why |
|---|---|---|---|---|
| SaaS MVP | Next.js | Node.js/Express | PostgreSQL | Fast iteration, full-stack JS |
| E-commerce | Next.js | Node.js or Python | PostgreSQL + Redis | SEO, caching, transactions |
| Mobile App | React Native | Node.js/Python | PostgreSQL | Cross-platform, shared logic |
| Real-time App | React | Node.js + WebSocket | PostgreSQL + Redis | Event-driven, low latency |
| Data Platform | React | Python/FastAPI | PostgreSQL + ClickHouse | Data processing, analytics |
| Enterprise | React | Java/Spring or .NET | PostgreSQL/Oracle | Stability, enterprise support |
| ML Product | React | Python/FastAPI | PostgreSQL + Vector DB | ML ecosystem, inference |
| Team Profile | Recommended Stack | Avoid |
|---|---|---|
| Full-stack JS | Next.js, Node.js, PostgreSQL | Go, Rust (learning curve) |
| Python Background | FastAPI, React, PostgreSQL | Heavy frontend frameworks |
| Enterprise Java | Spring Boot, React, PostgreSQL | Bleeding-edge tech |
| Startup (Speed) | Next.js, Supabase/Firebase | Complex microservices |
| Scale-Up | React, Go/Node, PostgreSQL | Monolithic frameworks |
| Framework | Best For | Learning Curve | Ecosystem | Hiring |
|---|---|---|---|---|
| React | Complex UIs, SPAs | Medium | Excellent | Easy |
| Next.js | Full-stack, SSR, SEO | Medium | Excellent | Easy |
| Vue.js | Simpler apps, gradual adoption | Easy | Good | Medium |
| Svelte | Performance-critical | Easy | Growing | Hard |
| Angular | Enterprise, large teams | Hard | Good | Medium |
Speed to MVP Long-term Maint Enterprise Ready
React ████████░░ ████████░░ █████████░
Vue █████████░ ███████░░ ██████░░░░
Angular ██████░░░░ █████████░ ██████████
| Framework | Language | Best For | Performance | Ecosystem |
|---|---|---|---|---|
| Express | Node.js | APIs, real-time | Good | Excellent |
| Fastify | Node.js | High-performance APIs | Excellent | Good |
| FastAPI | Python | ML APIs, async | Excellent | Good |
| Django | Python | Full-featured apps | Good | Excellent |
| Spring Boot | Java | Enterprise | Good | Excellent |
| Go (Gin/Echo) | Go | High performance | Excellent | Good |
| Rails | Ruby | Rapid prototyping | Moderate | Good |
| NestJS | TypeScript | Structured Node apps | Good | Good |
## Node.js (Express/Fastify/NestJS)
✅ Real-time applications (WebSocket)
✅ I/O-heavy workloads
✅ Full-stack JavaScript teams
✅ Microservices
❌ CPU-intensive tasks
❌ Heavy computation
## Python (FastAPI/Django)
✅ ML/Data Science integration
✅ Rapid prototyping
✅ Data processing pipelines
✅ Scientific computing
❌ High-concurrency I/O
❌ Real-time systems
## Go
✅ High-performance services
✅ System programming
✅ Concurrent workloads
✅ Microservices at scale
❌ Rapid prototyping
❌ Complex ORM needs
## Java (Spring Boot)
✅ Enterprise applications
✅ Complex business logic
✅ Transaction-heavy systems
✅ Large teams
❌ Quick MVPs
❌ Small projects
| Database | Type | Best For | Scale | Complexity |
|---|---|---|---|---|
| PostgreSQL | Relational | General purpose, ACID | High | Medium |
| MySQL | Relational | Web apps, read-heavy | High | Low |
| MongoDB | Document | Flexible schemas, JSON | High | Low |
| Redis | Key-Value | Caching, sessions | Very High | Low |
| Elasticsearch | Search | Full-text search | High | Medium |
| ClickHouse | Columnar | Analytics, time-series | Very High | Medium |
| DynamoDB | Key-Value | Serverless, AWS | Very High | Medium |
| Cassandra | Wide-column | Write-heavy, distributed | Very High | High |
Need ACID transactions?
├── YES → PostgreSQL
│
└── NO → What's your primary use case?
├── General purpose → PostgreSQL (still!)
├── Document storage → MongoDB
├── Caching → Redis
├── Search → Elasticsearch
├── Analytics → ClickHouse/BigQuery
├── Time-series → TimescaleDB/InfluxDB
└── Key-value at scale → DynamoDB/Cassandra
| Platform | Best For | Complexity | Cost |
|---|---|---|---|
| Vercel | Next.js, frontend | Very Low | $ - $$ |
| Railway | Simple deployments | Low | $ - $$ |
| Render | General apps | Low | $ - $$ |
| AWS | Everything, scale | High | $ - $$$$ |
| GCP | ML/Data, Kubernetes | High | $ - $$$$ |
| Azure | Enterprise, .NET | High | $ - $$$$ |
| DigitalOcean | Simple, affordable | Low | $ |
| Fly.io | Edge, global | Medium | $ - $$ |
┌──────────────────────────────────────────────────────────────────┐
│ MODERN SAAS STACK │
├──────────────────────────────────────────────────────────────────┤
│ │
│ FRONTEND BACKEND DATABASE │
│ ───────── ─────── ──────── │
│ Next.js 14 Node.js/Express PostgreSQL │
│ TypeScript TypeScript Prisma ORM │
│ Tailwind CSS REST/GraphQL Redis (cache) │
│ │
│ INFRASTRUCTURE AUTH PAYMENTS │
│ ────────────── ──── ──────── │
│ Vercel Clerk/Auth0 Stripe │
│ AWS S3 NextAuth Stripe Billing │
│ Cloudflare CDN │
│ │
│ MONITORING CI/CD ANALYTICS │
│ ────────── ───── ───────── │
│ Sentry GitHub Actions PostHog/Amplitude │
│ Datadog Vercel Preview Mixpanel │
│ │
└──────────────────────────────────────────────────────────────────┘
Best for: B2B SaaS, 0-1M users
Team size: 2-10 engineers
Time to MVP: 4-8 weeks
┌──────────────────────────────────────────────────────────────────┐
│ E-COMMERCE STACK │
├──────────────────────────────────────────────────────────────────┤
│ │
│ FRONTEND BACKEND DATABASE │
│ ───────── ─────── ──────── │
│ Next.js (SSR) Node.js/Python PostgreSQL │
│ TypeScript GraphQL/REST Redis │
│ Tailwind/Styled Medusa/Custom Elasticsearch │
│ │
│ PAYMENTS SHIPPING INVENTORY │
│ ──────── ──────── ───────── │
│ Stripe ShipStation Custom/ERP │
│ PayPal EasyPost Webhook sync │
│ │
│ CDN SEARCH QUEUE │
│ ─── ────── ───── │
│ CloudFront Algolia/Elastic SQS/BullMQ │
│ Cloudflare Typesense Redis │
│ │
└──────────────────────────────────────────────────────────────────┘
Best for: D2C, Marketplace
Team size: 5-20 engineers
Time to MVP: 8-16 weeks
┌──────────────────────────────────────────────────────────────────┐
│ ML PRODUCT STACK │
├──────────────────────────────────────────────────────────────────┤
│ │
│ FRONTEND API ML SERVING │
│ ───────── ─── ────────── │
│ React/Next.js FastAPI TorchServe/Triton │
│ TypeScript Python Docker/K8s │
│ Pydantic ONNX Runtime │
│ │
│ DATABASE VECTOR DB FEATURE STORE │
│ ──────── ───────── ───────────── │
│ PostgreSQL Pinecone Feast │
│ Redis Weaviate Redis │
│ pgvector │
│ │
│ ML OPS TRAINING MONITORING │
│ ───── ──────── ────────── │
│ MLflow SageMaker Weights & Biases │
│ Airflow Vertex AI Prometheus/Grafana │
│ │
└──────────────────────────────────────────────────────────────────┘
Best for: AI products, recommendation systems
Team size: 5-15 engineers + ML team
Time to MVP: 12-24 weeks
┌──────────────────────────────────────────────────────────────────┐
│ REAL-TIME STACK │
├──────────────────────────────────────────────────────────────────┤
│ │
│ FRONTEND BACKEND REAL-TIME │
│ ───────── ─────── ───────── │
│ React Node.js Socket.io │
│ TypeScript Express/Fastify WebSocket │
│ TypeScript Redis Pub/Sub │
│ │
│ DATABASE CACHE MESSAGE QUEUE │
│ ──────── ───── ───────────── │
│ PostgreSQL Redis Redis Streams │
│ Prisma In-memory Kafka (scale) │
│ │
│ PRESENCE STATE SYNC CONFLICT RESOLUTION │
│ ──────── ────────── ─────────────────── │
│ Redis CRDT/OT Yjs/Automerge │
│ Custom LiveBlocks Custom │
│ │
└──────────────────────────────────────────────────────────────────┘
Best for: Chat, collaboration, gaming
Team size: 5-15 engineers
Time to MVP: 8-16 weeks
| Factor | JavaScript/TS | Python | Go | Java | Rust |
|---|---|---|---|---|---|
| Learning Curve | Low | Low | Medium | Medium | High |
| Ecosystem | Excellent | Excellent | Good | Excellent | Growing |
| Performance | Good | Moderate | Excellent | Good | Excellent |
| Hiring Pool | Large | Large | Medium | Large | Small |
| Type Safety | TS: Good | Optional | Excellent | Excellent | Excellent |
| Memory Safety | GC | GC | GC | GC | Compile-time |
## Evaluation Checklist
1. **Team Expertise** (Weight: 30%)
- Current skills alignment?
- Learning curve acceptable?
- Training resources available?
2. **Project Requirements** (Weight: 30%)
- Performance requirements met?
- Feature set complete?
- Scalability path clear?
3. **Ecosystem** (Weight: 20%)
- Package availability?
- Community size?
- Third-party integrations?
4. **Long-term Viability** (Weight: 20%)
- Active maintenance?
- Corporate backing?
- Future roadmap?
| Anti-Pattern | Why It's Bad | Better Approach |
|---|---|---|
| Resume-Driven | Choosing tech for career, not project | Match to requirements |
| Hype-Driven | Picking latest without evaluation | Proven over trendy |
| Comfort-Only | Only familiar tech even when unsuitable | Evaluate objectively |
| Over-Engineering | Complex stack for simple needs | Start simple |
| Under-Engineering | Simple tools for complex needs | Plan for growth |
❌ "Let's use microservices from day one"
→ Start monolith, extract later
❌ "We need Kubernetes for our 3-person startup"
→ Use managed platforms (Vercel, Railway)
❌ "MongoDB because NoSQL is modern"
→ PostgreSQL handles 95% of use cases better
❌ "GraphQL for everything"
→ REST is simpler for most APIs
❌ "Let's build our own auth"
→ Use Auth0, Clerk, or established solutions
| Trigger | Action |
|---|---|
| Performance bottlenecks | Profile first, then consider |
| Team expertise mismatch | Train or hire before migrating |
| End of life/support | Plan 6-12 months ahead |
| Scale limitations | Validate limits with benchmarks |
| Security vulnerabilities | Patch if possible, migrate if not |
LOW RISK:
- Library/package updates
- Minor version upgrades
- Adding new services
MEDIUM RISK:
- Database version upgrades
- Framework major versions
- New deployment platform
HIGH RISK:
- Language/framework rewrites
- Database technology changes
- Monolith to microservices
| Project | Recommended Stack |
|---|---|
| Blog/CMS | Next.js + Headless CMS (Sanity/Contentful) |
| SaaS Dashboard | Next.js + Node.js + PostgreSQL |
| Mobile App | React Native + Node.js + PostgreSQL |
| E-commerce | Next.js + Medusa/Custom + PostgreSQL |
| Real-time Chat | React + Node.js + Socket.io + Redis |
| Data Dashboard | React + Python/FastAPI + PostgreSQL |
| ML Product | React + Python/FastAPI + PostgreSQL + Vector DB |
| API Service | Node.js or Python + PostgreSQL |
| Complexity | Description | Example Stack |
|---|---|---|
| Minimal | Single deployment, managed services | Vercel + Supabase |
| Simple | Separate frontend/backend | Vercel + Railway + PostgreSQL |
| Standard | Multiple services, caching | AWS ECS + RDS + Redis |
| Complex | Microservices, event-driven | K8s + Multiple DBs + Kafka |
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Detect common technical and organizational anti-patterns in proposals, architectures, and plans. Use when strategic-cto-mentor needs to identify red flags before they become problems.
日本語の概要は準備中です。原文の説明を表示しています。
Recommend architecture patterns (monolith, microservices, serverless, modular monolith) based on scale, team size, and constraints. Use when cto-architect needs to select the right architectural approach for a new system or migration.
日本語の概要は準備中です。原文の説明を表示しています。
Identify and challenge implicit assumptions in plans, proposals, and technical decisions. Use when strategic-cto-mentor needs to surface hidden assumptions and wishful thinking before they become costly mistakes.
日本語の概要は準備中です。原文の説明を表示しています。
Generate targeted clarifying questions (2-3 max) that challenge vague requirements and extract missing context. Use after request-analyzer identifies clarification needs, before routing to specialist agents. Helps cto-orchestrator avoid delegating unclear requirements.
日本語の概要は準備中です。原文の説明を表示しています。
Infrastructure and development cost estimation for technical projects. Use when planning budgets, evaluating build vs buy decisions, or projecting TCO for architecture choices.
日本語の概要は準備中です。原文の説明を表示しています。
Transform clarified user requests into structured delegation prompts optimized for specialist agents (cto-architect, strategic-cto-mentor, cv-ml-architect). Use after clarification is complete, before routing to specialist agents. Ensures agents receive complete context for effective work.
日本語の概要は準備中です。原文の説明を表示しています。