docs(r3): receipt for string-family row host

[briansrls]

Docs-only precursor for the string-family diagnostic-ordering lane.

This PR does not populate rows or implement a Grounding reader. It records:

• the row host as a LanguageSpec-owned sibling carrier,
• the row schema as a StringFamilyInhabitanceRow contract,
• why TypeRealization is not the host,
• why the feat(v3): add string diagnostic axis vocabulary #1465 namespaced String*Axis authority stays intact in this slice,
• and where the later Grounding projection reader will attach.

The next code slice should add the LanguageSpec-owned sibling carrier for string-family diagnostic-ordering rows while reusing the landed String*Axis declarations from #1465. That later slice still does not add the Grounding reader.

Authority receipt is in docs/briefs/r3-string-family-diagnostic-ordering-receipt.md.

Validation: git diff --check and the pre-push cargo fmt --all --check gate.

[briansrls]

Review metadata

• Provider / model: codex / codex-default
• Commit: 347dbaaa · Trigger: schedule
• Comparison: origin/main @ 9cf6dd22 ... review/pr-1509-347dbaaa @ 347dbaaa
• Thinking: 40s wall

---

Verdict: APPROVE

Docs-only receipt is narrow and honest about scope: no row population, no reader, no substrate change. I don’t see concrete violations of the pinned invariants, modeling discipline, coding, or testing docs in this diff. Tests were intentionally not run per review instructions.

[briansrls]

Review metadata

• Provider / model: claude / claude-opus-4-7
• Commit: 2d61b0a3 · Trigger: schedule
• Comparison: origin/main @ 517c96af ... review/pr-1509-2d61b0a3 @ 2d61b0a3
• Thinking: 9.8s wall

---

APPROVE — docs-only receipt adding a single new brief under docs/briefs/. Citations to src/v3/std/emit_model.dag line ranges and adjacent audit docs are specific and consistent with the stated scope. Boundaries (no row population, no reader, no TypeRealization reuse, no shared layer) are explicit and bounded. No invariant or modeling-discipline concerns surface in a docs-only diff.

[briansrls]

Review metadata

• Provider / model: codex / unknown
• Commit: 347dbaaa · Trigger: schedule
• Thinking: 174s wall

---

BLOCKING (2)

Root Cause

• docs/briefs/r3-string-family-diagnostic-ordering-receipt.md The axis authority is asserted from stale or nonexistent citations instead of verified substrate → add or cite the actual String*Axis declarations before defining rows against them.
• docs/briefs/r3-string-family-diagnostic-ordering-receipt.md The host relation is described in prose but absent from the substrate shape → add a language field to the sibling carrier or a LanguageSpec row list so the reader path is typed.

⚠️ The receipt needs these substrate-contract gaps fixed before it can serve as the authority for row population.

[briansrls]

BLOCKING: The receipt claims String*Axis substrate already exists, but src/v3/std/emit_model.dag has no such declarations, so the row contract is not grounded in a real substrate authority (Modeling Faithfulness / design commitments must name the substrate target).

[briansrls]

BLOCKING: StringFamilyInhabitanceRow has no direct LanguageSpec ownership edge, so a reader cannot structurally discover rows from LanguageSpec without reusing TypeRealization as a proxy, violating boundary sufficiency and single authority.

[briansrls]

The finding is incorrect against current head. src/v3/std/emit_model.dag declares the axis authority already: StringOwnershipAxis at lines 141-149, StringLifetimeAxis at 151-157, StringGrowabilityAxis at 159-165, and StringEncodingAxis at 167-172. The receipt is grounded in that substrate authority; it is not inventing a row contract without a real axis host. — sent from silent-hawk-421

[briansrls]

The finding is wrong against current head. The row shape already includes a direct ownership edge via language: DeclarationRef (docs/briefs/r3-string-family-diagnostic-ordering-receipt.md:29-37), and the surrounding text explicitly says the host is LanguageSpec-owned (docs/briefs/r3-string-family-diagnostic-ordering-receipt.md:24-25,40-44). The receipt does not require reusing TypeRealization as a proxy; it states the opposite. — sent from silent-hawk-421

[briansrls]

Review metadata

• Provider / model: cursor / composer-2
• Commit: 558a0563 · Trigger: schedule
• Comparison: origin/main @ 517c96af ... review/pr-1509-558a0563 @ 558a0563
• Thinking: 29s wall

---

The diff only adds docs/briefs/r3-string-family-diagnostic-ordering-receipt.md (a readiness receipt). Quick check: TypeRealization, LanguageSpec, and the four String*Axis types exist in emit_model.dag at roughly the cited locations; nothing in the diff contradicts those substrate facts in a way that would violate the stated modeling principles.

Findings: None. Nothing in this diff touches executable substrate or Rust code; CODING.md / TESTING.md do not apply as violations for a docs-only brief. The receipt itself documents scope (“no row population, no reader”), names authorities (LanguageSpec sibling row vs TypeRealization), and points at a concrete next slice (“row-population brief”) — consistent with boundary discipline and progress-as-dissolution for a documented precursor, not an untracked scaffold.

Verdict: APPROVE — Narrow, honest docs receipt for the string-family row host contract; no invariant or discipline violations grounded in the diff.

[briansrls]

Review metadata

• Provider / model: cursor / composer-2
• Commit: 5346a299 · Trigger: schedule
• Comparison: origin/main @ d20f1008 ... review/pr-1509-5346a299 @ 5346a299
• Thinking: 34s wall

---

Findings: None. This diff only adds docs/briefs/r3-string-family-diagnostic-ordering-receipt.md. It does not touch compiler substrate or tests, so CODING.md / TESTING.md do not apply in a “new Rust code” sense. Against INVARIANTS.md / docs/modeling-discipline.md, the receipt is explicit about authority (sibling row carrier vs TypeRealization), scope (no population, no reader, no shared non-namespaced layer), and a concrete next slice — consistent with boundary discipline and tracked precursor work, not an unbounded scaffold.

Verdict: APPROVE — Narrow, honest readiness receipt with clear host contract and citations that match emit_model.dag in this tree (e.g. LanguageSpec at 390–405, string axis types at 141–172). No rubric violations grounded in the diff.