About
First Report tracks how news stories spread: who published first, who followed, what was copied and what was changed later. It also collects facts about each outlet, with sources.
The site is built from two parts: feed, which reads outlets' feeds and groups headlines into stories, and ledger, which stores facts about outlets. Each page shows when it was last built.
What the numbers mean
Story
Headlines that feed grouped as the same event because their text is similar. The grouping is automatic and sometimes wrong.
First
The headline with the earliest publish time in its feed. Feeds sometimes get times wrong, so we also show when we picked it up.
Outlets, items
How many outlets covered a story, and how many headlines they published. No outlet counts for more than another.
Reach
How the front page is ordered. It counts outlets that wrote their own version of the story, adds one for each extra language, then breaks ties by recent coverage and by time. Stories only one outlet covered are listed separately. The topic and the outlets involved play no part. Full rules: docs/algorithms.md.
Copy
The same text published by more than one outlet. Copies don't count as separate coverage. Wire stories are meant to be copied.
Revised, withdrawn
The headline or summary changed after publication, or the item disappeared from its feed.
Claim
A fact about an outlet from one source, with a link, the licence and the date we read it. When sources disagree we show both.
Stale
Older than the source's refresh interval, so it may be out of date.
Measurement
Facts we measured ourselves from domain records, the outlet's home page and its feed. Each one gives the time window and counts it's based on.
What First Report doesn't do
- First Report has no score or rating of its own. The State, Sanctioned and Deprecated tags each come from a named source, shown on hover.
- It doesn't summarise stories. Headlines are the outlets' own.
- It only reads feeds, never full article pages. The raw fetches are kept.
- It doesn't filter outlets by quality. Every feed in the pool is treated the same.
Sources
Where our facts about outlets come from. If you reuse them, follow the source's licence.
For machine readers
Every page has index.json and index.md beside it with with the same content. /llms.txt states the rules of use; /sitemap.xml lists every page; /search/index.json is the index of outlets and stories. Please cite the original source and keep its licence. Don't turn these facts into a score.
Built from an export at 2026-11-21 10:16 UTC covering 96 outlets in the pool and 96 outlets with claims. First Report is open source: github.com/h0ds/kagami.