What Spis is#

Spis is an evidence-grade reference corpus: 700 records across 13 interface families, each carrying its source, hashes, provenance class, and measured state.

A record exists only with its evidence. A missing observation is recorded as an evidence gap and is never promoted into prose.

Words this documentation uses#

`record` — one measured reference: media, source URL, hashes, provenance class, and status.
`provenance class` — how the media was obtained: `local-product-run`, `local-browser-run`, or `upstream-owner-media`.
`evidence gap` — something the corpus does not know, recorded instead of guessed.
`catalog` — one family of records, such as web-app-examples or cli-examples.