start-it

start-it is a guided CLI for scaffolding opinionated project baselines across backend, frontend, AI/ML, and DSA workflows.

Instead of asking users to pick from a flat template list, the CLI now walks through:

1. app type
2. implemented stack
3. stack-specific options
4. project metadata
5. deterministic scaffold generation

Current Coverage

Backend

• Node.js + TypeScript + Express
• NestJS
• Python + FastAPI

Frontend

• React + Vite
• Next.js

AI / ML

• Python + FastAPI Serving
• R Analytics Pipeline
• C++ Inference Utility

DSA-specific

• C++
• Python

What The CLI Collects

The prompt flow is stack-aware.

• Backend: databases, security preset, logging, monitoring, testing
• Frontend: routing, styling, optional shadcn/ui, state, data fetching, testing
• AI / ML: serving or runtime mode, packaging, tracking, validation, logging, testing
• DSA-specific: track, runner style, verification mode

Installation

npm install -g start-it-cli

Or run it without installing:

npx start-it-cli

Usage

start-it-cli

The CLI then guides the project setup interactively.

For the full command reference, flags, output-directory options, and non-interactive examples, see USAGE.md.

Example Flows

Backend Example

App type: Backend
Stack: Node.js + TypeScript + Express
Databases: PostgreSQL, Redis
Security preset: bcrypt + JWT
Logging: Pino
Monitoring: Prometheus-ready
Testing: Jest + Supertest

Frontend Example

App type: Frontend
Stack: React + Vite
Routing: React Router
Styling: Tailwind CSS
UI add-on: shadcn/ui starter
State: Zustand
Data fetching: TanStack Query
Testing: Vitest + React Testing Library

AI / ML Example

App type: AI / ML
Stack: Python + FastAPI Serving
Serving mode: Realtime + batch endpoints
Packaging: MLflow-ready
Tracking: MLflow
Validation: Pydantic + Pandera
Testing: Pytest + HTTPX

DSA Example

App type: DSA-specific
Stack: Python
Track: Interview preparation
Runner style: Function-first
Verification: Pytest

Generated Project Shape

Generation is deterministic and stack-specific.

• Backend projects generate service-ready API scaffolds
• Frontend projects start from a provider-style baseline and are then customized
• AI / ML projects generate serving, analytics, or inference workspaces
• DSA projects generate practice workspaces with sample problems and runner/test setup

Each generated project also includes:

• .cursorrules
• docs/AGENTS.md
• docs/instructions.md
• stack-specific README.md

Development

Setup

npm install
npm run build

Run Locally

npm run dev

Test

npm test

Developer Docs

• Contribution guide: CONTRIBUTING.md
• Local development: docs/DEVELOPMENT.md
• Architecture: docs/ARCHITECTURE.md
• CI / CD: docs/CI_CD.md

Project Structure

src/
├── cli.ts
├── generator.ts
├── workflow.ts
├── types.ts
├── templates/
├── frontend/
├── aiml/
├── dsa/
└── __tests__/

Notes

• Only implemented stacks are shown in the workflow
• Some older static templates still exist in the repository, but the main CLI path is now app-type driven
• Frontend generation prefers provider-style baselines when applicable, with deterministic post-processing afterward

License

MIT