Skip to content

About

AI-powered feed builder that discovers sources, filters noise with configurable pipelines, clusters passed articles into stories, and serves the results through a simple web UI and RSS feeds.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Repository files navigation

What it does

Feed Builder lets you create custom news feeds from multiple source types (RSS, YouTube, Reddit, Twitter/Nitter, Tavily web search). Each feed consists of a list of sources and filters.

Key features

  • Multi-source feeds — aggregate RSS feeds, YouTube channels, Reddit, Twitter/Nitter, and web search results into a single feed
  • AI filtering pipeline — build a pipeline of blocks (keyword filters, LLM scoring, deduplication, relevance checks) that each article must pass through
  • Stories view — articles are clustered into stories, giving a digest-style view of what's happening
  • Pipeline versioning — every pipeline change is versioned; revert to any previous version in one click
  • Replay — re-evaluate all historical articles against an updated pipeline without re-fetching
  • Audits — run a full AI audit of a feed's performance over a time window; get a structured report on pass rates, source quality, and pipeline recommendations, with a proposed pipeline diff you can review and apply
  • Manual verdicts — override the pipeline's pass/filter decision on individual articles
  • RSS output — every feed exposes a standard RSS endpoint for use in any reader
  • Demo mode — deploy a read-only frontend where all mutating actions are disabled

Environment variables

See .env.example for the full list. Required:

  • DATABASE_URL — Postgres connection string (or SQLite path for dev)
  • ANTHROPIC_API_KEY — used for all pipeline LLM calls
  • FRONTEND_URL — comma-separated list of allowed frontend origins for CORS

Optional:

  • OPENAI_API_KEY, YOUTUBE_API_KEY, TAVILY_API_KEY, NITTER_BASE, proxy credentials

About

AI-powered feed builder that discovers sources, filters noise with configurable pipelines, clusters passed articles into stories, and serves the results through a simple web UI and RSS feeds.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages