The AskNews platform Storylines
Track a storyline, not a feed of articles
Storylines follows what the world is covering as it develops. Each update reports what changed, where sources disagree and how coverage is moving across outlets, countries and languages, as structured data your agents can act on.
Widely-covered storylines, selected by breadth of coverage rather than clicks.
One storyline, one ID
What changed
Where sources disagree
Outlets, countries and languages
Coverage and sentiment over time
Each update is written from at least five different outlets
What's happening, and how it got there
Alerts tell you when something you defined happens. Storylines are the record of what the world is covering: AskNews clusters its news corpus into persistent storylines, and new coverage joins the storyline it belongs to as an update. One ID follows a storyline from its first report through every development, for as long as it keeps developing.
Every update, written from a diverse set of sources
The articles behind each update are chosen for diversity: at most one per outlet, sampled in proportion to the storyline's mix of source countries, so no single outlet or country sets the narrative. An update is only written when those articles span at least five different outlets, and its citations trace each claim back to the article it came from.
Analytics on every update, not just a summary
Each update separates what is new from the background and lists the claims its sources disagree on. It carries its outlets, source countries and languages, a series of coverage and sentiment across the storyline's updates, the people and organizations involved and how they connect, and the Reddit discussion where there is one.
What your agents can do with it
- List the storylines the world is covering without writing a query, including the ones nobody thought to set an alert for.
- Keep a storyline's ID and read only the updates added since the last check.
- Check a claim against the sources that disagree on it before repeating it.
- Feed coverage and sentiment series into research pipelines, dashboards or forecasting features.
What each update returns
- What is new, the key takeaways and where sources contradict each other.
- Outlets, source countries and languages, with coverage and sentiment across the storyline's updates.
- Entities and their relationships, citations back to each article, and Reddit discussion where it exists.
- A confidence score from the model's token log probabilities: a measure of certainty, not of truth.
MCP, CLI and SDK quickstart
One ID, every development
Search and fetch storylines through the API, the Python and TypeScript SDKs or the CLI, filtered by category, region, time window or query. Store a storyline's ID and fetch it again for its latest updates. The WebUI shows the same storylines and their analytics if you would rather browse first.
Prefer to try it in the browser first?
The WebUI runs the same capability without an integration, so you can judge the output before you write any code.
Give your agents the record behind the headlines
Start with the storylines in play today, then keep the ones that matter and read only what changes.