Epigram offers an AI-powered platform for consuming news in short, digestible formats, aiming to reduce information overload. It targets users seeking efficient and personalized news consumption, providing AI-generated summaries and deeper analysis from trusted sources.

How It Works

Epigram leverages a Next.js frontend with AI capabilities powered by OpenAI's GPT-4o-mini and Exa for news summarization. News data is sourced via the Mediastack API. User preferences drive a personalized feed, and AI is used for both concise summaries and in-depth analysis. Caching and rate limiting are handled by Upstash Redis.

Quick Start & Requirements

• Install: git clone https://github.com/panda-sandeep/epigram
• Prerequisites: Node.js 20+, npm/yarn, Git. Requires API keys for Mediastack, Exa, OpenAI, and Upstash Redis.
• Setup: Copy .env.example to .env.local and populate with API keys and secrets.
• Run: npm install, npm run dev
• Populate Data: curl --header "X-Epigram-Secret: <your-secret>" http://localhost:3000/api/news/populate
• Docs: Try Epigram.news »

Highlighted Details

• AI-powered summaries and "deep dives" for news articles.
• Personalized news feed based on user interests.
• Responsive design for cross-device accessibility.
• Serverless deployment options (Google Cloud Run, Vercel, Netlify).

Maintenance & Community

• Project led by Sandeep Panda and Fazle.
• Contributions are welcomed via pull requests.
• Announcement available on X: https://x.com/Sandeepg33k/status/1875373269307814113

Licensing & Compatibility

• Licensed under AGPL-3.0. This license is copyleft and may require derivative works to also be open-sourced under AGPL-3.0, potentially impacting commercial use or integration into closed-source projects.

Limitations & Caveats

The project requires multiple third-party API keys, including paid services like OpenAI and Mediastack, which represent a significant operational cost and dependency. The AGPL-3.0 license may impose restrictions on commercial use or integration into proprietary software.