Off — your saved articles stay on this device only.
Your sync code--------
Enter this code on your other devices to keep saved articles in sync.
No saved articles yet. Tap the ♡ on any card to save it.
Sort:
Building story clusters...
Insights
Live Stats
—
Articles
—
Flagged
—
Clusters
—
Flag rate
Flag Breakdown
Outlet Scorecards (today)
Article Analysis
Filter articles
🔥 Hot
Lean
Flags
How Brous News Works
What it does
Brous News aggregates headlines from 15 outlets across the political spectrum — from Breitbart (R) to The Guardian (LC) — and uses rule-based pattern matching to flag narrative techniques commonly used to bias coverage.
Narrative flags
ANON
Anonymous sourcing — "sources say," "people familiar," "insiders told." Protects legitimate whistleblowers but also enables unverifiable claims.
LOADED
Loaded language — politically charged words like "radical," "woke," "deep state," "regime." Signals ideological framing over neutral reporting.
UNNAMED
Unnamed experts — "experts say," "scientists warn," "officials claim" without naming them. Lends authority without accountability.
SPECUL
Speculation as news — "could signal," "fears grow," "some wonder." Presents guesswork with the tone of fact.
EMOTIVE
Emotive language — "shocking," "bombshell," "slams," "destroys." Designed to provoke emotional reactions rather than inform.
FRAMING
Selective framing — "admits," "caught lying," "silenced by." Presupposes guilt or conspiracy before evidence is presented.
Divergence score
Measures how differently outlets frame the same story. Calculated from the spread of political leans covering it plus variance in flag counts. High divergence (≥3) means left and right outlets are telling meaningfully different versions of the same event.
Outlet scorecards
Tracks each outlet's flag rate for today's articles. A consistently high flag rate doesn't mean the outlet is wrong — it means their language choices match patterns associated with narrative bias. Use it as one signal among many.
Limitations
All analysis is pattern-matching only — no AI, no editorial judgment. Flags can produce false positives (e.g. "experts say" in a legitimate science story). The goal is to make you notice the language, not to tell you what to believe.