about

sitrep is a front page for defence and tech news. it polls around a hundred rss feeds (specialist outlets, wires, and official newsrooms) and groups items that cover the same story into a single ranked entry. the more outlets pick a story up, and the faster they do it, the higher it sits.

who broke it

every story credits the earliest item. broke defense news 09:12 means defense news published first. when the earliest item came from a company or government newsroom, the story is marked with its origin instead, so you can always see when coverage started life as a press release.

how clustering works

items are matched by classic text similarity (tf-idf over headline and summary terms, cosine distance). no machine learning models, no language models, no ai anywhere in the pipeline. grouping and ranking are plain arithmetic over who published what, and when.

headlines only

sitrep stores headlines, feed summaries, thumbnails, and links. it never stores article bodies. each headline opens a short story page with the full coverage timeline, and every outlet on that page links straight to the source, with tracking parameters stripped and the destination never rewritten. reading happens on the publisher's site.

takedowns

if you run one of the polled feeds and want your content removed or a correction made, email benjnib09@gmail.com and it will be handled promptly.