Throughline Monitor · Narrative monitoring
See the talking points as they spread.
Throughline Monitor watches the outlets that push a narrative, classifies each story by the talking point it echoes, and shows you the framing — side by side, day by day. The recurring thread running through the coverage, surfaced while it still matters. It tracks published content, never people.
Why the framing wins
Propaganda doesn’t arrive labelled.
A talking point lands in one outlet, then another, then ten — each wording it a little differently, all pushing the same frame. By the time you notice the pattern, it has already done its work. Spotting a framing is a skill; doing it continuously, across dozens of outlets, in another language, is a full-time job nobody has time for.
Throughline Monitor does that watching for you — and shows you the thread, so you can recognise the frame before it wins.
How it works · a live instance
Built first to track war propaganda about Ukraine.
The same engine, pointed at one of the clearest cases there is. It is running now — a calm, public-interest reference plus a live tracker.
Watch the outlets that push the narrative
It follows a defined set of openly state-affiliated, institutional outlets by RSS — not individuals, not community pages. The live build watches Russian state media for coverage of the war in Ukraine.
Classify each story by the talking point it echoes
Every item is matched to a recurring narrative — “denazification”, “it’s a proxy war”, “both sides are as bad” — then deduped and rolled into a daily digest. Each match is a hint for human review, never a verdict.
See the framing gap, side by side
It places independent, on-the-ground reporting beside the state-media version of the same event, so the difference in framing is visible at a glance — with a 30-second verify routine to pressure-test any claim.
Principled by design
It watches narratives, not people.
It does
- Track published content and framing from openly-affiliated institutional outlets.
- Classify each item by the recurring narrative it echoes.
- Surface the framing difference, side by side.
- Roll everything up into a calm daily digest.
It never
- Collect, store, or rank data about any person.
- Track accounts, communities, ethnicity, or nationality.
- Submit reports to anyone, automatically or otherwise.
- Encourage naming, shaming, or vigilantism.
It focuses on conduct and content — published news and narratives — never on who someone is or where they come from. Matching is a hint for human interpretation, and nothing is reported anywhere automatically.
Point it at your narrative
One war was just the first.
Anywhere institutional outlets push a coordinated frame, the same engine works. Define the sources and the narratives that matter to you; Throughline Monitor watches and surfaces the thread.
Newsrooms & journalists
Catch a coordinated talking point the moment it spreads across affiliated outlets.
Fact-checkers & researchers
Classify a disinformation campaign by the framings it recycles — and export the receipts.
Diaspora & advocacy groups
See how state media is framing your community — content and conduct, never people.
Election-integrity & democracy orgs
Track foreign-influence narratives across a campaign, beside independent coverage.
Media-literacy educators
Teach narrative recognition with live, real-world examples and a verify routine.
OSINT & policy analysts
Monitor a defined set of institutional sources and roll it into a daily digest.
Why it works
Watching plus classification plus contrast.
A keyword feed just gives you more to read. Throughline Monitor pairs tireless source-watching with narrative classification and a side-by-side framing view — so instead of more noise, you get the pattern: which story, which frame, how it differs from the independent account. It keeps a human in the loop by design.
Put a throughline on the narrative that matters to you.
Define the outlets and the framings worth watching, and see the thread surface day by day — content and conduct, never people.
Early access. Access isn’t open yet — the button is a placeholder for now.