What data does Cumulative Web Inc expose to AI systems?
Cumulative Web Inc exposes its public catalog, knowledge graph, search and licensing surfaces to AI systems — and nothing else. Every response carries X-CWI-License headers; commercial AI training requires a license.
What's public
llms.txt — the map of every public surface. /catalog.json — the track catalog with Spotify links. /graph.json — the knowledge graph. /kit.json — the AI learning kit. /query?q= — scored track search, top 5. /license — machine-readable licensing terms. /changes?since= — crawl deltas. /.well-known/agent-card.json — the machine identity. cumulativeweb.com/data/*.json — roster, artist and producer JSON. sanqa/catalog.json — the SANQA machine-readable catalog. The public receipts feed and RadioStation JSON-LD round it out.
What's not
No credentials, no internal logs, no infrastructure internals are exposed — the public edge serves catalog, search and licensing data by design. Requests hit the edge at metadata level only; request bodies are never logged.