wildlifechorus

wildlifechorus / condenseit

Public

Self-hosted AI news digest. Collect RSS feeds, YouTube channels, website diffs, Google News searches, Hacker News, Reddit, GitHub Releases, and podcasts, summarize with a local LLM (Ollama) or OpenRouter, learn your preferences from star ratings and engagement signals, and read a daily digest in the browser.

18
0
89% credibility
Found May 22, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

CondenseIt is a self-hosted personal news digest that automatically collects content from sources you choose (RSS feeds, YouTube channels, websites, podcasts, Google News, Hacker News, Reddit, and more), uses AI to summarize everything, learns your preferences from how you interact with it, and delivers a ranked daily digest to your browser. It works entirely on your own computer or on a server you control, with no dependence on third-party services. The more you read and rate articles, the better it becomes at surfacing content you'll actually care about.

How It Works

1
📰 You decide you want your own news digest

You're tired of scrolling through generic news apps. You want articles that actually matter to you, curated and summarized in one place.

2
⚙️ You set up your news sources

You pick the blogs, YouTube channels, podcasts, and websites you care about. You can even add Google News searches for topics you follow.

3
You choose how AI does the work
💻
Local AI (free)

Run everything on your Mac using free, local AI models. No internet needed for the smart stuff.

☁️
Cloud AI

Let cloud AI do the heavy lifting. You just need an account with a AI service provider.

4
✨ Your first digest appears

The app reads all your sources, summarizes every article, and ranks them by what matters to you. You see a beautiful digest in your browser.

5
⭐ You start rating articles

As you read, you give stars to articles you love and skip ones you don't. The more you rate, the smarter the rankings become.

6
📈 The system learns your taste

After just a handful of ratings, the engine starts boosting topics, categories, and sources you prefer. You can watch your preference profile grow.

🎉 You have your perfect daily digest

Every morning, your personalized digest waits for you. Articles are ranked by your actual interests, not algorithms you don't control. You can even set it to run automatically.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is condenseit?

CondenseIt is a self-hosted AI news digest that pulls content from RSS feeds, YouTube channels, websites, Google News, Hacker News, Reddit, GitHub Releases, and podcasts, then summarizes everything using either a local LLM (Ollama with Metal GPU support) or OpenRouter's cloud API. The system learns your preferences from star ratings and engagement signals, ranking articles by relevance to your interests. You read the results in a browser-based digest with filtering, search, and a full admin panel for managing sources and settings. It's built in Python with a Preact frontend, runs behind a password, and can be deployed on a Mac, VPS, or via Docker.

Why is it gaining traction?

The preference learning engine is the hook. Instead of manually configuring filters, CondenseIt watches what you star, save, and dismiss, building a profile that improves over time. The ranking transparency ("Why ranked here?") shows exactly what drove each article's position. Budget controls prevent runaway API costs, and the dual LLM mode (free local via Ollama, paid cloud via OpenRouter) lets you start without spending a cent. The cold-start bootstrap feature seeds the engine with a plain-text description of your interests before any ratings exist.

Who should use this?

Developers and technical users who want a private, customizable news aggregator without surrendering data to third-party services. Security researchers tracking CVEs, open-source maintainers monitoring release feeds, or anyone drowning in RSS subscriptions will get the most value. The local Ollama mode works well for Mac users with Metal GPUs; the VPS deployment suits those wanting a shared digest accessible from any device.

Verdict

CondenseIt is a polished concept with solid documentation and multiple deployment paths, but the 18-star count signals early-stage software. The credibility score of 0.8999999761581421% reflects this limited community validation. If you want a self-hosted alternative to commercial news aggregators and don't mind a project still finding its footing, this is worth exploring. Start with local Ollama mode to avoid costs, and expect to configure sources manually until the UI matures.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.