Kickoff v3.0.0

Modern CLI scaffolding tool for production-ready projects with 2025 tech stacks. Multi-language support (TypeScript, Python, Go, Rust, Deno), AI/ML integrations, and security best practices baked in.

Features

Multi-Language: TypeScript/Node, Python, Go, Rust, Deno support
2025 Tech Stacks: Supabase, Neon, Turso, Hono, TanStack Start, and more
AI/ML Ready: Vector databases, embeddings, local AI (Ollama, MLX for Apple Silicon)
Modern Auth: Clerk, Kinde, Auth0, Better Auth, Supabase Auth
14 Presets: One-command project setup for common use cases
Interactive Docs: HTML documentation accessible via CLI

Installation

npm install -g project-kickoff

Or run directly with npx:

npx project-kickoff create my-app

Quick Start

# Create a new project with interactive wizard
kickoff create my-app

# Use a preset for instant setup
kickoff create my-app --preset saas-starter

# List all available presets
kickoff list

# Dry run to see what would be created
kickoff create my-app --dry-run

Documentation

# Open quick start guide in browser
kickoff docs

# Browse all presets with examples
kickoff presets

# Explore 2025 tech stacks
kickoff stacks

Presets (One-Command Setup)

Preset	Stack	Description
`saas-starter`	Next.js + Supabase + Drizzle	Production SaaS with auth, DB, design system
`tanstack-hono`	TanStack Start + Turso + Better Auth	Type-safe full-stack with edge DB
`edge-api`	Hono + Turso + Bun	Blazing fast edge API
`api-microservice`	Hono + Neon + Drizzle	Serverless microservice
`fastapi-starter`	FastAPI + PostgreSQL + SQLAlchemy	Python API with modern tooling
`python-ml-api`	FastAPI + pgvector + LangChain	ML-ready Python API
`go-microservice`	Gin + PostgreSQL + GORM	Go microservice
`rust-api`	Axum + PostgreSQL + SQLx	Rust API with compile-time safety
`ai-rag-app`	Next.js + Supabase Vector + Vercel AI	RAG application
`ai-agent`	Hono + Ollama + LangChain	Local AI agent
`mlx-local`	FastAPI + MLX + Chroma	Apple Silicon optimized AI
`quick-cli`	Commander + TypeScript	CLI tool
`landing-page`	HTML/CSS/JS + nginx	Static site
`mcp-tool`	TypeScript + stdio/SSE	MCP server for AI tools

Project Types

Full-Stack JavaScript/TypeScript

Type	Framework	Description	Generated Files
`nextjs`	Next.js 15	App Router, RSC, Server Actions	`app/`, `components/`, `lib/`, `next.config.ts`
`tanstack-start`	TanStack Start	React 19, type-safe routing	`app/`, `routes/`, `tanstack.config.ts`

Frontend

Type	Framework	Description	Generated Files
`vite-react`	Vite + React 19	SPA with TypeScript	`src/`, `vite.config.ts`, `index.html`
`static`	HTML/CSS/JS	Static site with nginx	`public/`, `nginx.conf`

Backend - Node/Bun

Type	Framework	Runtime	Generated Files
`hono-api`	Hono	Bun/Node	`src/index.ts`, `src/routes/`, `src/middleware/`
`elysia-api`	Elysia	Bun	`src/index.ts`, `src/routes/`, Eden types
`express-api`	Express	Node	`src/index.ts`, `src/routes/`, `src/middleware/`

Backend - Deno

Type	Framework	Description	Generated Files
`fresh-api`	Fresh	Islands architecture	`routes/`, `islands/`, `fresh.config.ts`

Backend - Python

Type	Framework	Description	Generated Files
`fastapi`	FastAPI	Async, auto-docs	`app/`, `main.py`, `requirements.txt`/`pyproject.toml`
`litestar`	Litestar	High-performance	`app/`, `main.py`, `requirements.txt`

Backend - Go

Type	Framework	Description	Generated Files
`gin-api`	Gin	Fast, popular	`main.go`, `handlers/`, `models/`, `go.mod`
`fiber-api`	Fiber	Express-like	`main.go`, `handlers/`, `go.mod`
`echo-api`	Echo	Minimalist	`main.go`, `handlers/`, `go.mod`

Backend - Rust

Type	Framework	Description	Generated Files
`axum-api`	Axum	Tokio-based	`src/main.rs`, `src/handlers/`, `Cargo.toml`
`actix-api`	Actix	Actor-based, fastest	`src/main.rs`, `src/handlers/`, `Cargo.toml`

Tools

Type	Description	Generated Files
`cli`	Commander CLI	`src/index.ts`, `src/commands/`, optional prompts/config
`mcp-server`	MCP server	`src/index.ts`, `src/tools/`, stdio/SSE transport
`library`	npm package	`src/index.ts`, `tsup.config.ts`, vitest/jest
`worker`	Background jobs	`src/worker.ts`, BullMQ + Redis

Database Providers

Serverless (JavaScript/TypeScript recommended)

Provider	Type	Features	Required Keys	Monthly Cost
`supabase`	PostgreSQL	Auth + Storage + Realtime	`SUPABASE_URL`, `SUPABASE_ANON_KEY`	Free tier / $25+
`neon`	PostgreSQL	Branching, edge, scale-to-zero	`DATABASE_URL`	Free tier / $19+
`turso`	SQLite (libSQL)	Edge-first, global replication	`TURSO_DATABASE_URL`, `TURSO_AUTH_TOKEN`	Free tier / $29+
`convex`	Reactive BaaS	Real-time, type-safe	`CONVEX_DEPLOYMENT`	Free tier / $25+

BaaS (Backend-as-a-Service)

Provider	Type	Features	Required Keys	Monthly Cost
`pocketbase`	SQLite BaaS	Single file, auth, realtime	None (self-hosted)	Free (self-hosted)
`firebase`	NoSQL	Google BaaS, auth, storage	`FIREBASE_*` config	Free tier / Pay-as-you-go

Self-Hosted (Docker required)

Provider	Type	Docker Image	Memory	Generated Files
`postgres-local`	PostgreSQL	`postgres:16-alpine`	256MB+	`docker-compose.yml`, `init.sql`
`mysql-local`	MySQL	`mysql:8`	512MB+	`docker-compose.yml`, `init.sql`
`mongodb-local`	MongoDB	`mongo:7`	512MB+	`docker-compose.yml`
`sqlite`	SQLite	None	N/A	`db/` directory

Enterprise

Provider	Type	Features	Required Keys	Monthly Cost
`planetscale`	MySQL + Vitess	Branching, serverless	`DATABASE_URL`	Free tier / $29+
`cockroachdb`	Distributed SQL	Multi-region, ACID	`DATABASE_URL`	Free tier / $0.50/vCPU-hr

ORM Choices

JavaScript/TypeScript

ORM	Style	Features	Generated Files
`drizzle`	SQL-like	Fast, lightweight, type-safe	`src/db/schema.ts`, `drizzle.config.ts`
`prisma`	Schema-first	Great DX, migrations	`prisma/schema.prisma`, `src/db/client.ts`
`kysely`	Query builder	Type-safe SQL	`src/db/kysely.ts`

Python

ORM	Style	Features	Generated Files
`sqlalchemy`	Standard	Mature, flexible	`app/models.py`, `app/database.py`
`sqlmodel`	Pydantic	FastAPI creator's ORM	`app/models.py`
`tortoise`	Async	Django-like, async-first	`app/models.py`

Go

ORM	Style	Features	Generated Files
`gorm`	Full ORM	Auto-migrations, hooks	`models/`, `database/db.go`
`sqlx-go`	Raw SQL	Type-safe, no ORM magic	`database/queries.go`

Rust

ORM	Style	Features	Generated Files
`sqlx-rust`	Compile-time	Async, checked queries	`src/db.rs`, `migrations/`
`sea-orm`	Active Record	Async, built on SQLx	`src/entities/`, `src/db.rs`
`diesel`	Sync	Mature, type-safe	`src/schema.rs`, `diesel.toml`

Authentication Providers

Hosted (Easy Setup)

Provider	Features	Required Keys	Monthly Cost
`clerk`	Pre-built components, social login	`CLERK_SECRET_KEY`, `NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY`	Free tier / $25+
`kinde`	B2B-focused, SAML SSO, feature flags	`KINDE_*` config	Free tier / $25+
`auth0`	Enterprise-grade, Okta-owned	`AUTH0_*` config	Free tier / $23+
`workos`	Enterprise SSO, directory sync	`WORKOS_*` config	Free tier / Pay-as-you-go

Self-Hosted

Provider	Features	Required Keys	Generated Files
`better-auth`	Modern, great DX	`BETTER_AUTH_SECRET`	`src/lib/auth.ts`
`lucia`	Lightweight, minimal	`LUCIA_*` config	`src/lib/auth/`
`authjs`	NextAuth successor	`AUTH_SECRET`, `AUTH_*`	`auth.ts`, `middleware.ts`

Platform-Specific

Provider	Platform	Features	Required Keys
`supabase-auth`	Supabase	Built-in with DB	Supabase keys
`convex-auth`	Convex	Built-in with DB	Convex keys
`firebase-auth`	Firebase	Built-in with DB	Firebase config
`pocketbase-auth`	PocketBase	Built-in with DB	None

AI/ML Features

AI Frameworks

Framework	Language	Features	Required Keys	Generated Files
`vercel-ai`	TypeScript	React hooks, streaming	`OPENAI_API_KEY` or other	`src/lib/ai.ts`, `app/api/chat/`
`langchain`	Python/JS	Chains, agents, RAG	Model provider keys	`src/lib/langchain/`
`llamaindex`	Python/JS	Data framework, RAG	Model provider keys	`src/lib/llamaindex/`
`mastra`	TypeScript	TypeScript-first	Model provider keys	`src/lib/mastra/`
`instructor`	Python/JS	Structured outputs	Model provider keys	`src/lib/instructor/`

Vector Databases

Provider	Type	Features	Required Keys	Monthly Cost
`pinecone`	Managed	Scalable, production-ready	`PINECONE_API_KEY`	Free tier / $70+
`supabase-vector`	pgvector	Built into Supabase	Supabase keys	Included
`pgvector`	PostgreSQL ext	Use existing Postgres	None	Included with DB
`qdrant`	Self-hosted	Rust-based, fast	None	Free (self-hosted)
`chroma`	Self-hosted	Lightweight, dev-friendly	None	Free (self-hosted)
`weaviate`	Self-hosted	GraphQL, hybrid search	None	Free (self-hosted)
`turbopuffer`	Managed	Serverless, edge	`TURBOPUFFER_API_KEY`	Pay-as-you-go
`milvus`	Self-hosted	Enterprise, distributed	None	Free (self-hosted)

Embedding Providers

Provider	Models	Required Keys	Cost per 1M tokens
`openai`	text-embedding-3-small/large	`OPENAI_API_KEY`	$0.02 / $0.13
`voyage`	voyage-large-2, code-optimized	`VOYAGE_API_KEY`	$0.10+
`cohere`	embed-english-v3	`COHERE_API_KEY`	$0.10
`google`	text-embedding-004	`GOOGLE_API_KEY`	$0.025
`together`	Various	`TOGETHER_API_KEY`	$0.008
`ollama`	nomic-embed, mxbai	None (local)	Free
`huggingface`	sentence-transformers	None (local)	Free
`fastembed`	Lightweight	None (local)	Free

Local AI/LLM Providers

Provider	Platform	Features	Memory	GPU Required
`ollama`	Cross-platform	Easy setup, many models	8GB+	Optional
`lmstudio`	Desktop	GUI + server mode	8GB+	Optional
`jan`	Desktop	Desktop app + API	8GB+	Optional
`mlx`	Apple Silicon	MLX framework, fast on Mac	16GB+	No (uses Neural Engine)
`mlx-lm`	Apple Silicon	MLX LLM package	16GB+	No
`vllm`	Linux/CUDA	Production serving	16GB+	Yes (CUDA)
`localai`	Docker	OpenAI drop-in replacement	8GB+	Optional
`tgi`	Docker	HuggingFace inference	16GB+	Yes (CUDA)
`llamacpp`	Cross-platform	Low-level, max control	8GB+	Optional
`text-gen-webui`	Docker	Full-featured UI	16GB+	Yes (recommended)

Web Servers / Reverse Proxy

Server	Features	Config File	Use Case
`caddy`	Auto-HTTPS, simple config	`Caddyfile`	Recommended for most projects
`nginx`	High-performance, battle-tested	`nginx.conf`	High-traffic, complex routing
`traefik`	Docker/K8s native, auto-discovery	`traefik.yml`	Container orchestration

Runtime Options

Runtime	Version	Best For	Package Manager
`node`	20+	Compatibility, ecosystem	npm/pnpm/yarn
`bun`	1.0+	Speed, all-in-one	bun
`deno`	2.0+	Security, web standards	deno
`python`	3.11+	ML/AI, data science	pip/poetry/uv
`go`	1.21+	Performance, simplicity	go mod
`rust`	stable	Safety, performance	cargo

Python Package Managers

Manager	Features	Generated Files
`uv`	Fast, Rust-based (recommended)	`pyproject.toml`, `uv.lock`
`poetry`	Dependency management + publishing	`pyproject.toml`, `poetry.lock`
`pip`	Standard, battle-tested	`requirements.txt`
`pipenv`	Pipfile + lock file	`Pipfile`, `Pipfile.lock`

Server Requirements & Cost Guide

Development Machine

Configuration	RAM	CPU	Storage	Cost
Minimal (Node/Bun)	8GB	2 cores	10GB	-
Standard (with Docker)	16GB	4 cores	50GB	-
AI/ML (local models)	32GB+	8 cores	100GB+	-
Apple Silicon AI (MLX)	16GB+ unified	M1/M2/M3/M4	50GB+	-

Production VPS (Monthly)

Provider	Config	RAM	CPU	Cost	Best For
Hetzner	CX22	4GB	2 vCPU	$4.50	Basic APIs
Hetzner	CX32	8GB	4 vCPU	$8	Standard apps
Hetzner	CX42	16GB	8 vCPU	$16	Heavy workloads
Railway	Starter	8GB	8 vCPU	$5+	Serverless
Fly.io	Small	1GB	1 shared	$5	Edge deployment
DigitalOcean	Basic	4GB	2 vCPU	$24	General purpose

GPU Cloud (for AI inference)

Provider	GPU	RAM	Cost/hour	Best For
RunPod	RTX 3090	24GB	$0.39	Development
Lambda Labs	A10	24GB	$0.75	Production
Vast.ai	Various	Varies	$0.20+	Budget

Required User Data

Always Required

Project name: kebab-case, max 50 characters
GitHub username: For deployment workflows

Conditional Requirements

Feature	Required Data	Environment Variables
Supabase	Supabase project	`SUPABASE_URL`, `SUPABASE_ANON_KEY`
Neon	Neon project	`DATABASE_URL`
Turso	Turso database	`TURSO_DATABASE_URL`, `TURSO_AUTH_TOKEN`
Clerk	Clerk app	`CLERK_SECRET_KEY`, `NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY`
Kinde	Kinde app	`KINDE_ISSUER_URL`, `KINDE_CLIENT_ID`, `KINDE_CLIENT_SECRET`
Auth0	Auth0 tenant	`AUTH0_SECRET`, `AUTH0_BASE_URL`, `AUTH0_ISSUER_BASE_URL`, `AUTH0_CLIENT_ID`, `AUTH0_CLIENT_SECRET`
OpenAI	API key	`OPENAI_API_KEY`
Pinecone	API key + index	`PINECONE_API_KEY`, `PINECONE_INDEX`
Production domain	Domain name	`DOMAIN` (for Traefik/Caddy)
Go projects	Module path	`go.mod` path
Rust projects	Crate name	`Cargo.toml`

Files Generated Per Option

Base Files (All Projects)

project/
├── CLAUDE.md           # AI coding assistant context
├── README.md           # Project documentation
├── .gitignore          # Git ignore rules
├── .env.example        # Environment template
└── .github/
    └── workflows/
        └── deploy.yml  # CI/CD pipeline

With Docker Deployment

project/
├── Dockerfile          # Multi-stage, secure build
├── docker-compose.yml  # With resource limits
└── .dockerignore       # Docker ignore

With Database

project/
├── src/db/
│   ├── schema.ts       # Drizzle/Prisma schema
│   └── client.ts       # Database client
└── drizzle.config.ts   # or prisma/schema.prisma

With Auth

project/
├── src/lib/auth/
│   ├── index.ts        # Auth configuration
│   └── middleware.ts   # Auth middleware
└── middleware.ts       # Route protection

With AI/ML

project/
├── src/lib/ai/
│   ├── index.ts        # AI client setup
│   ├── embeddings.ts   # Embedding functions
│   └── chains/         # LangChain chains (if applicable)
└── .env.example        # With AI API keys

AI-Enhanced Stack Validation

Kickoff includes intelligent stack validation that catches compatibility issues before you start coding.

Dual-Mode Validation

Mode	Speed	Requires API Key	Best For
Rule-based	Instant	No	Known incompatibilities, common mistakes
LLM-enhanced	2-5s	Yes	Architecture advice, edge cases, recommendations

Usage

# Rule-based validation only (default, always runs)
kickoff create my-app

# Enable AI-enhanced validation
kickoff create my-app --validate

# Choose specific LLM provider
kickoff create my-app --validate --provider anthropic
kickoff create my-app --validate --provider openai
kickoff create my-app --validate --provider gemini

Supported LLM Providers

Provider	Model	Environment Variable
Anthropic	Claude Sonnet 4.5	`ANTHROPIC_API_KEY`
OpenAI	GPT-5	`OPENAI_API_KEY`
Google	Gemini 3 Flash	`GEMINI_API_KEY`

Set any of these in your .env file. The validator auto-detects available providers and falls back gracefully.

What Gets Validated

Rule-based checks:

D1 + Prisma incompatibility (edge runtime issues)
NoSQL databases with SQL ORMs
Platform-specific auth requiring matching database
Apple Silicon-only options (MLX) on other platforms

AI-enhanced analysis:

Architecture recommendations for your stack
Performance considerations (e.g., Turbopuffer cold query latency)
Better alternatives based on your use case
Edge case compatibility issues

Example Output

Stack Validation Results
──────────────────────────────────────────────────
📋 Rule-Based Validation:
  ✅ No compatibility issues found

🤖 AI-Enhanced Validation (Claude - 92% confidence):
  ⚠️ Warning: Better-Auth + Next.js + Bun
     Build failures reported with this combination
     💡 Consider using Node.js runtime or Clerk for auth

  💡 Recommendations:
  • Consider adding rate limiting for production API
  • Supabase RLS pairs well with your auth choice
──────────────────────────────────────────────────
✅ Validation PASSED with warnings - Review recommendations

Security Features

Based on production incident lessons:

Never build on VPS: GitHub Actions builds only
Non-root containers: All Dockerfiles use USER appuser (uid 1001)
Resource limits: CPU/memory limits in docker-compose
Security headers: Traefik/Caddy middleware for HSTS, CSP, XSS
Health endpoints: /api/health or /health for monitoring
Secrets management: .env.example templates, no hardcoded secrets
Security scanning: npm audit, Trivy in CI/CD

Testing

# Run all tests
npm test

# Watch mode
npm run test:watch

# Coverage report
npm run test:coverage

Development

# Clone
git clone https://github.com/abe238/project-kickoff.git
cd project-kickoff

# Install
npm install

# Development
npm run dev

# Build
npm run build

# Test
npm test

License

ISC

Built with lessons learned from production deployments, security incidents, and real-world AI/ML applications.

About the Builder

Hola. I'm Abe Diaz. By day, I'm Sr. Manager of Disaster Relief at Amazon, leading a team that uses AI and logistics to deliver 26+ million relief items and response technology across 200+ disasters worldwide. On weekends, I build artisanal developer tools like this :-P

Other Open Source Projects

Project	Description
gg-deploy	Domain → GitHub Pages in 60 seconds
gemini-deep-research	CLI for Gemini's Deep Research Agent
aipm-resume-analyzer	Resume analysis against AI PM frameworks

Name		Name	Last commit message	Last commit date
Latest commit History 22 Commits
.taskmaster		.taskmaster
docs		docs
src		src
.env.example		.env.example
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
package-lock.json		package-lock.json
package.json		package.json
tsconfig.json		tsconfig.json
vitest.config.ts		vitest.config.ts

Folders and files

Latest commit

History

Repository files navigation

Kickoff v3.0.0

Features

Installation

Quick Start

Documentation

Presets (One-Command Setup)

Project Types

Full-Stack JavaScript/TypeScript

Frontend

Backend - Node/Bun

Backend - Deno

Backend - Python

Backend - Go

Backend - Rust

Tools

Database Providers

Serverless (JavaScript/TypeScript recommended)

BaaS (Backend-as-a-Service)

Self-Hosted (Docker required)

Enterprise

ORM Choices

JavaScript/TypeScript

Python

Go

Rust

Authentication Providers

Hosted (Easy Setup)

Self-Hosted

Platform-Specific

AI/ML Features

AI Frameworks

Vector Databases

Embedding Providers

Local AI/LLM Providers

Web Servers / Reverse Proxy

Runtime Options

Python Package Managers

Server Requirements & Cost Guide

Development Machine

Production VPS (Monthly)

GPU Cloud (for AI inference)

Required User Data

Always Required

Conditional Requirements

Files Generated Per Option

Base Files (All Projects)

With Docker Deployment

With Database

With Auth

With AI/ML

AI-Enhanced Stack Validation

Dual-Mode Validation

Usage

Supported LLM Providers

What Gets Validated

Example Output

Security Features

Testing

Development

License

About the Builder

Other Open Source Projects

About

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors 2

Languages

Packages