shape-doctor.py: manifest kinds counts a file that declares a registered kind validated wherever it sits in a root that runs that kind's check, though each check reads only fixed paths, so a contracts/project.yaml in an assembly root, a contracts/family.yaml or contracts/spec-pin.yaml in a family holder, and a contracts/<x>-pin.yaml that no leg or declared descent names are ok and COMPLIANT though nothing reads them #439

Lane: openreposhape-3 (openRepoShape-3)

What happens. #375 (#327) made the row ask whether a kind's check runs on this kind of root (ManifestKind.runs_in). It does not ask whether that check reads the file's path, and each gate reads fixed paths. So every other contracts/*.yaml that declares one of those kinds is still counted validated, and the row says "every declared kind has a registered validator".

Measured on main 45e01f4 with the suite's own fixtures: a scaffolded Atlas root and a holder made by family.py init + add, each COMPLIANT before the change. Each file below was added and committed by itself:

assembly root + contracts/project.yaml      (kind: project-manifest, id: "007", nothing else)
  ok   manifest kinds   6 manifest(s), every declared kind has a registered validator
  COMPLIANT   (exit 0)        detail.validated: [..., "contracts/project.yaml (project-manifest)", ...]

assembly root + contracts/release-pin.yaml  (kind: pinned_contract_manifest, commit: not-a-commit)
  ok   manifest kinds   6 manifest(s), every declared kind has a registered validator
  COMPLIANT   (exit 0)        detail.validated: [..., "contracts/release-pin.yaml (pinned_contract_manifest)", ...]

assembly root + contracts/naming-old.yaml   (kind: repository-naming-policy, forms: 7)
  ok   manifest kinds   6 manifest(s), every declared kind has a registered validator
  COMPLIANT   (exit 0)

family holder + contracts/family.yaml       (kind: family-manifest, one member row with no gitlink, commit: "not-a-commit")
  ok   manifest kinds  4 manifest(s), every declared kind has a registered validator
  COMPLIANT   (exit 0)        detail.validated: [..., "contracts/family.yaml (family-manifest)", ...]

family holder + contracts/spec-pin.yaml     (kind: pinned_contract_manifest, commit: not-a-commit)
  ok   manifest kinds  4 manifest(s), every declared kind has a registered validator
  COMPLIANT   (exit 0)

None of these files reaches a gate. detail.registered_elsewhere is absent in all five, because each kind's check does run on that kind of root.

Where.

• shape-doctor.py:check_manifest_kinds (:1674). The decision is at :1728-1731: elif ctx.kind not in entry.runs_in(), otherwise known.append. The ok sentence is at :1772-1774.
• shape-doctor.py:manifest_candidates (:1372): the two top-level manifests and every contracts/*.yaml.
• shape-doctor.py:ManifestKind.runs_in (:748) and MANIFEST_VALIDATORS (:756-777).

What each gate reads, on the same sha:

Why it matters. The row's sentence is a claim about this root, and for these files it is false: no gate reads them. That is #211's claim and #327's again, one level down: the row decides by root kind when it should decide by path. The pin cases are the realistic ones. A neutral pin whose neutral_product_pins: entry was dropped, or a leg pin left in a holder, reads as checked.

Direction: count a file validated only at a path its check reads on this root. Those paths are:

• the root's own manifest;
• contracts/shape-pin.yaml;
• the declared leg and neutral-product pins;
• contracts/repository-naming.yaml.

Report any other file that declares a registered kind apart from validated, and say that no gate reads it. #375 ruled registered_elsewhere a FINDING because a file that declares kind: asserts what it is. The writer should follow that or argue for a note.

The NEXT takes the file out as git holds it (#375's removal_ways), or the human declares it. It never declares a pin itself: AGENTS.md says "Do NOT declare a pin the human has not asked for". It never suggests renaming the file onto a declared path.

What must be kept:

• On an untouched root, every file that is validated today stays so, and the row stays byte-identical.
• shape-doctor.py calls an assembly root whose project.yaml does not parse (a merge conflict) NOT A SHAPE ROOT, exit 2, and its NEXT line tells the reader to adopt that assembly root in place while the Triad row calls it a single repository #205's skip, shape-doctor.py reads only project.yaml at a root that also carries a family.yaml, so an assembly root whose family.yaml declares two unmounted members is COMPLIANT, and its manifest kinds row says that kind has a registered validator, which is never run #211's family_beside and shape-doctor.py: manifest kinds counts a registered kind validated only on a kind of root that runs its check, so a contracts/family.yaml in an assembly root, or a contracts/project.yaml in a family holder, is a FINDING naming the git rm that takes it out instead of ok and COMPLIANT (#327) #375's registered_elsewhere are unchanged.

Open PR #358 (#291) adds declared_pins, which is the set this needs. Whichever of the two lands second merges check_manifest_kinds again. The path-classification-policy entry, which the registry covers with this row itself, is outside this issue.

Source. PR #375 (#327), Left out: "a registered kind at a path its check never reads, in a root where that check does run, is still counted validated. Examples are a contracts/project.yaml in an assembly root, a contracts/family.yaml in a holder, and a contracts/<x>-pin.yaml that no leg or declared descent names." PR #375's review also found that "a holder with contracts/family.yaml is still ok, 4 manifests".

Related:

• shape-doctor.py reads only project.yaml at a root that also carries a family.yaml, so an assembly root whose family.yaml declares two unmounted members is COMPLIANT, and its manifest kinds row says that kind has a registered validator, which is never run #211 and shape-doctor.py: manifest kinds counts a contracts/*.yaml that declares a registered kind as validated though that kind's validator never runs on this kind of root, so a contracts/family.yaml in an assembly root, or a contracts/project.yaml in a family holder, is ok and COMPLIANT where the same bytes at the root are a FINDING #327 (closed): the same claim at the root's top and by root kind.
• PR shape-doctor.py: a family.yaml beside an assembly root's project.yaml is a manifest kinds FINDING, INVALID (manifest-kinds), naming each member it pins with no gitlink, off its pin or with no 40-hex pin and each gitlink under members/ no row names, so a root its family gate refuses is no longer COMPLIANT, exit 0, with that file counted as validated (#211) #220's Left out, which stated the root-kind rule.
• shape-doctor.py: the manifest kinds row's NEXT for a file that does not load (a conflicted or non-UTF-8 contracts/shape-pin.yaml) is "register the kind in shape-doctor.py's MANIFEST_VALIDATORS ... or remove the file from the root", so the one fix the doctor prints for a damaged pin is to delete it #291 and PR shape-doctor.py: the manifest kinds row puts back a shape or leg pin under contracts/ that does not load with the git command pin_restore picks from its own state, shows a copy the shape pin has a row for or a neutral product pin with a read-only git diff, and leaves any other file to a person whatever it is called, so its NEXT no longer tells the reader to delete the pin (#291) #358 (open; the same row, for files that do not load). shape-doctor.py: the manifest kinds row puts back a shape or leg pin under contracts/ that does not load with the git command pin_restore picks from its own state, shows a copy the shape pin has a row for or a neutral product pin with a read-only git diff, and leaves any other file to a person whatever it is called, so its NEXT no longer tells the reader to delete the pin (#291) #358 already treats "a contracts/release-pin.yaml that no leg and no product names" as the project's own file.