- Audited branch: main (
<sha>) - Types:
<from registry, or resolved> - Verdict: operational | not operational
- Date:
<YYYY-MM-DD>
develop vs main: ahead <n>, behind <n>. Classify by content, not commit count: promotion merge commits leave main permanently ahead with identical trees (benign, inherent to merge-commit promotions - a cherry-pick would be an empty no-op). A drift finding only when main carries content develop lacks; the audit's content-based branch check is the authority.
| Dimension | Letter | Intent | Verdict | Evidence (file:line) |
|---|---|---|---|---|
| csharp | ||||
| nuget | ||||
| pypi | ||||
| python | ||||
| dotnet-publish | ||||
| hugo | ||||
| docker | ||||
| branch-model | ||||
| carried-scope | ||||
| repo-setup | ||||
| runtime-secrets | ||||
| verbatim-tree | ||||
| linter-parity | ||||
| recurring-violations | ||||
| readme-structure | ||||
| workflow (WORKFLOW.md 5A/5B) |
Verdict values: operational | not-operational | N/A. Letter and Intent values: pass | miss | N/A, where a letter miss with intent pass is a drift finding and both missing is a defect. Mark a row N/A, in all three columns, only when its checks govern a construct the repo does not contain, and never delete a row, so the report shows the dimension was judged.
<defect>- input/condition -> observed vs expected, atfile:line.
<letter miss, intent satisfied>-file:line.
<e.g. resolve classificationPending: types = [...]>