A platform that roasts Valorant players based on their stats.
WHIFF.GG fetches Valorant player statistics from tracker.gg and generates AI-powered roasts based on performance. Built with Next.js, combining web scraping and AI generation with a Valorant-themed UI.
- Player stats scraping from Valorant tracker using Puppeteer
- AI-powered roast generation via GitHub Models API (GPT-4o-mini)
- Player validation for public accounts
- Comprehensive stats display (rank, K/D, win rate, etc.)
- Headless browser automation
- User enters Valorant player ID (e.g.,
Player#TAG) - App validates player existence and public profile status
- Scrapes detailed statistics from tracker.gg
- Sends stats to GitHub Models API for roast generation
- Displays roast and stats in Valorant-themed interface
- Frontend: Next.js 16, React, TailwindCSS, shadcn/ui
- Scraping: Puppeteer with Stealth Plugin
- AI: GitHub Models API (GPT-4o-mini)
This project only works in local development environments.
Valorant tracker.gg actively blocks requests from cloud/production environments like vercel.
- Node.js 18+
- GitHub Personal Access Token with access to GitHub Models
- Clone the repository:
git clone https://github.com/Starz099/whiff.gg.git
cd whiff.gg- Install dependencies:
npm install- Create a
.envfile in the root directory:
GITHUB_AI_TOKEN=your_github_token_here- Run the development server:
npm run dev
# or
yarn dev
# or
pnpm dev
# or
bun dev- Open http://localhost:3000 with your browser.
GITHUB_AI_TOKEN: GitHub Personal Access Token with access to GitHub Models API
POST /api/roast: Validates player, fetches stats, and generates roast- Body:
{ userId: "PlayerName#TAG" } - Returns:
{ message: string, stats: object }
- Body:
Contributions are welcome. Open an issue or submit a pull request for improvements or bug fixes.
This project is for educational purposes. Please respect tracker.gg's terms of service.