This page shows which component writes and which reads each of the six contracts, summarising living-doc-utilities'
Documentation contracts, where every
contract and rule is defined. It restates no rule.
Read before: Data flows · Next: Error codes
Each file's record roots, and the roles behind each writer and reader: Artifact rules — Contracts. Where each file lands: Data flows — Files.
| Contract | Artifact | Written by | Read by |
|---|---|---|---|
doc-source-v1.0.0 |
doc-source.json |
collector-gh, mode doc-source |
toolkit normalize-docs and coverage-matrix |
ui-tests-v1.0.0 |
ui-tests.json |
collector-gh, mode ui-tests |
toolkit normalize-tests and coverage-matrix |
generator-ready-v1.0.0 |
generator-ready.json |
toolkit normalize-docs |
generator-markdown, document type technical-project |
ui-test-catalog-v1.0.0 |
ui-test-catalog.json |
toolkit normalize-tests |
generator-markdown, document type ui-test-catalog |
coverage-matrix-v1.0.0 |
coverage-matrix.json |
toolkit coverage-matrix |
generator-markdown, document type coverage-matrix |
doc-entities-v1.0.0 |
doc-entities.json |
planned: collector-gh mode doc-issues, collector-ad |
planned: toolkit |
Every file carries one metadata object, the same model in all six contracts. It records the tool that wrote the
file and its living-doc-utilities version, the CI run, the project and the systems the data came from, when the file
was written, and the file's own counts. A toolkit output also lists the files it was made from. Its fields:
The metadata envelope.
entity_id: the id authored in an entity's header, such asUS-001, and the only key the files join on (Entity identity).source_ref: where the entity was read, with a link back to it; nothing joins on it (Entity identity).state_origin: whether a person wrote the entity's state or the pipeline derived it, as for every Feature (State and state origin).
Every contract rule, R1 to R13 and the named rules, is listed one line each in Documentation contracts — Rules, and each line links the page that defines it.