Hysen Labshackathon-starter
A boilerplate for Node.js web applications
Hackathon Starter bundles the boring parts of a web app
Built to skip the slow start of a hackathon, this Node.js boilerplate ships with OAuth logins, password resets, 2FA, file upload, and a growing set of API and AI examples.
Why it exists
The README opens with the familiar hackathon grind: decide what to build, pick a programming language, pick a web framework, pick a CSS framework, and only then can teammates start contributing. It even calls out the hours you can lose on something as plain as Sign in with Facebook if OAuth 2.0 is unfamiliar. The author says the primary focus was simplicity and ease of use, making the project as generic and reusable as possible, and that in the worst case it works as a learning guide.
The authentication stack
Authentication is the deep end of this boilerplate. Local accounts support email and password, passwordless, and passkey or biometrics. OAuth 2.0 covers Google, Microsoft, Facebook, LinkedIn, X (Twitter), Twitch, GitHub, and Discord. Around that sits user profile and account management, Gravatar, password management for change, reset, and forgot flows, email verification, and two factor authentication with email codes and authenticator apps. You can link multiple OAuth providers to one account, revoke tokens, and delete the account.
Beyond logins
The rest of the feature list is broader. There is a contact form powered by SMTP through Mailgun, AWS SES, and similar, file upload, and device camera support. The UI adds flash notifications, a dark mode toggle that defaults to OS preference, reCAPTCHA and rate limit protection, CSRF protection, an MVC project structure, Node.js clusters, HTTPS proxy support via ngrok or Cloudflare, Sass stylesheets, Bootstrap 5.3, and a go to production checklist.
API and AI examples
Newer additions lean on AI. The README mentions an AI agent example using ReAct with tool calling, MongoDB session persistence, and input guardrails, plus RAG with semantic and embedding caching, and text and image LLMs including GPT-OSS and Qwen3.6. Model providers run through LangChain, Groq, and Hugging Face. The API examples span payments like Stripe, PayPal, and QuickBooks, maps, stocks, social platforms, and more.
Editorial conclusion
The author frames it as simple and generic on purpose: if you only need Google sign in, it still works as a learning guide. Two testimonials in the README report years of reuse and hackathon prizes.
Community notes