Site map: what pages exist
Four groups. The pairs people mix up are “entity / event” (same layout, different subject) and “graph preview / full graph” (same data, different view).
Find content
Start here: what people are rehearsing, or how a subject / event is being read.
Every public rehearsal, grouped by topic. Multiple takes on one question sit side by side.
All rehearsals on one topic, plus the lead rehearsal’s full report.
Events and entities ranked by source heat. Fetched, not hand-edited.
Fetched events grouped by type (launch / regulation / funding…), paginated.
The two detail pages (easiest to confuse)
Both are “everything about one thing”. The only difference is whether the subject is an actor or a news event — the layout is identical.
About one actor: overview, related entities, related events, timeline, probability factors, AI causal chain, graph preview.
Same layout as the entity page, but the subject is “what happened”, with the sources that reported it.
The two graph views
Two views of the same data: the preview is a quick look, the full page is draggable / filterable.
Do it yourself
The sandbox is the entry; a watch is it running for you; a report is the takeaway.
Ask a question, pick 1–3 focus areas, run a multi-round rehearsal.
Saved rehearsals: rerun only when new sources land, and tell you what changed.
A downloadable artifact: overview, graph, timeline, probability, sources — exportable.
Every rehearsal whose window closed, including the ones we got wrong.
The machine-readable list is /sitemap.xml (for search engines). This page is for people.