QuickWiki

A simple web app that quickly summarizes Wikipedia articles using AI. The frontend is static (in public/), styles are built with Tailwind CSS, and the backend endpoint runs on Vercel in api/summarize.js.

Features

• Searches Wikipedia and selects the best matching article
• Fetches a clean plaintext extract of the article
• Generates a short AI summary (configurable number of sentences)

Requirements

• Node.js 18+
• Vercel account (for deployment)
• Groq API key in environment variable GROQ_API_KEY

Installation

npm install

CSS build (Tailwind)

• One-off build:

npm run build:css

• Watch during development:

npm run build:css:watch

• Minified build:

npm run build:css:minify

Input: src/input.css Output: public/output.css

Local development

The frontend is static. Open public/index.html in your browser after generating public/output.css. To serve locally you can use, for example:

npx serve public

The API endpoint is designed for Vercel (/api/summarize). To run locally with Vercel CLI:

npx vercel dev

Remember to set the environment variable:

export GROQ_API_KEY="<your_key>"

API

• GET /api/summarize?topic=<query>&length=<sentence_count>
  • topic (required): article/topic name on Wikipedia
  • length (optional): number of sentences in the summary (default 3)

Example (with Vercel or vercel dev running):

curl "http://localhost:3000/api/summarize?topic=Ada%20Lovelace&length=3"

License

Licensed under the MIT License