About starfeed.com — who runs it and how it works

What starfeed.com is

• •An aggregator: about twenty collectors (news RSS, Google/Bing/Naver news, Wikidata, Wikipedia, YouTube, Reddit, Mastodon, Bluesky, GDELT, MusicBrainz and others) run around the clock and store headlines with their original link, outlet and publication time.
• •A curated layer of roughly 4,600 stars that people actually look for. Only these pages are offered to search engines; the much larger archive stays browsable but is marked noindex.
• •A writing layer: for curated stars we publish daily briefings written by an AI model from the collected sources only. Every sentence carries numbered source references and a date. Briefings that fail automatic checks are not published.

What starfeed.com is not

• •Not a newsroom. We do not report, interview or verify events ourselves. Our contribution is collection, translation, summary and transparent labelling.
• •Not a rumour site. Claims reported by a single outlet are marked as such. Speculative wording is rejected by the automatic checks.
• •Not a place that copies articles. We show headlines and short summaries and send readers to the outlet that did the reporting.

Who writes what

• •Machine translation of headlines is labelled, and readers can flag a bad translation; two flags hide it and show the original.
• •Fan translations, schedule tips, aliases and corrections are labelled as fan contributions and can be reported; two reports hide them pending review.
• •AI briefings are labelled “AI summary from cited sources”. When a person has reviewed one it is labelled “reviewed by a person”.

Operator and contact

starfeed.com is run independently and is not affiliated with any company. The fastest way to reach us about a specific page is the “Request a correction” button on that page, which lands in the operator queue. Public channel: @starfeedcom on X.

Our editorial principles and corrections policy are published separately and linked from every content page.