Repolode
Submit
Tiendil

Tiendil/feeds.fun

Self-hosted news reader that automatically tags feed entries with LLMs and lets you score and filter news by those tags.

Open-source alternative to:Feedly

View on GitHubWebsite
Stars
397
Forks
19
Open issues
67
Last commit
16 days ago

What is it?

Feeds Fun is a news reader that assigns tags to every feed entry, then uses user-defined rules to score news by those tags. Readers can exclude or require tags, sort by score or date, and mark items as read, so only the most relevant articles remain.

Tags are produced by tag processors. The main one, llm_general, asks OpenAI or Gemini models to detect tags, and other processors handle simpler cases such as domain names, tags supplied by the feed, and uppercase titles. Raw tags are then cleaned by normalizers that merge variants like "a-book-review" and split comparisons like "google-vs-microsoft". Third-party models that expose an OpenAI-compatible or Gemini-compatible API can also be used.

The project runs as a Python backend (an ASGI API plus loader and librarian workers) with a frontend that is built as static content. It supports single-user and multi-user setups via Docker, and includes plugins for sources such as Reddit, GitHub, YouTube, ArXiv and Hacker News.

Key features

  • 01Automatically assigns tags to every news entry using LLMs or simpler processors.
  • 02Scores news with rules based on tags.
  • 03Filters news by including or excluding tags.
  • 04Sorts news by score, date and other criteria.
  • 05Tracks which items you have already read.
  • 06Supports single-user and multi-user Docker deployments.
  • 07Includes source-specific plugins for Reddit, GitHub, YouTube, ArXiv and Hacker News.

Use cases

  • Keeping up with many RSS feeds while reading only the most important items.
  • Self-hosting a personal news filter that uses your own OpenAI or Gemini key.
  • Building tag-based views over news from technical sources such as Hacker News, arXiv and GitHub.

Quick start

  1. 1

    Install the backend package from PyPI

    pip install ffun
  2. 2

    Download the SpaCy model used by the librarian worker

    python -m spacy download en_core_web_lg
  3. 3

    Run the database migrations (requires PostgreSQL configured via FFUN_POSTGRESQL__* variables)

    ffun migrate
  4. 4

    Start the API server

    uvicorn ffun.application.application:app --host 0.0.0.0 --port 8000 --workers 1

Things to know

  • Needs a PostgreSQL database and a minimal configuration before it runs.
  • Full tag generation requires an OpenAI or Gemini API key (or a compatible third-party model).
  • The README does not state a clear license, so check the LICENSE file before reuse.

In collections

RSS readers and feed toolsSelf-hosted apps

Topics

#rss-reader#news-aggregator#llm#tagging#self-hosted#feed-reader#python#feed#feed-aggregator#large-language-models

FAQ

What is feeds.fun?
Self-hosted news reader that automatically tags feed entries with LLMs and lets you score and filter news by those tags. Feeds Fun is a news reader that assigns tags to every feed entry, then uses user-defined rules to score news by those tags. Readers can exclude or require tags, sort by score or date, and mark items as read, so only the most relevant articles remain.
Is feeds.fun free to use? What license does it have?
feeds.fun is open source under the Other license. Check the license terms before using it in commercial products.
Is feeds.fun still maintained?
Actively maintained. The last commit to feeds.fun was on Oct 22, 2026.
What can I use feeds.fun for?
Keeping up with many RSS feeds while reading only the most important items. Self-hosting a personal news filter that uses your own OpenAI or Gemini key. Building tag-based views over news from technical sources such as Hacker News, arXiv and GitHub.
How do I install feeds.fun?
Install the backend package from PyPI: pip install ffun Download the SpaCy model used by the librarian worker: python -m spacy download en_core_web_lg Run the database migrations (requires PostgreSQL configured via FFUN_POSTGRESQL__* variables): ffun migrate Start the API server: uvicorn ffun.application.application:app --host 0.0.0.0 --port 8000 --workers 1
Where can I get feeds.fun?
The source code and installation instructions are on GitHub: https://github.com/Tiendil/feeds.fun

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects