test(v3): ValueBody substrate vs Rust mirror audit (E8 / #1531)

[briansrls]

Summary

Adds three ValueBody mirror audit tests to existing m2_substrate_inhabitance_test.rs (alongside the existing sum_variants(…, "ValueBody") ratchet): parses the type ValueBody sum in src/v3/std/substrate.dag from disk (include_str!), compares to an expected constructor list, and uses an exhaustive match over dag::ValueBody sample instances so new Rust variants fail compile until the audit is updated. A third test documents the known 3 (substrate) vs 5 (Rust runtime) gap and names the missing generation surface (extend reflected substrate.dag + regen when Scalar / List / Map carriers are generated from the Rust mirror — see dag.rs ValueBody docs).

No ValueBody shape edits, no parser/lowering changes, no substrate.dag edits in this PR.

---

Debt receipt (#1532)

Disposition

Detail

Debt paid

Disposition #1 — ValueBody Rust↔.dag mirror / isomorphism ratchet (R3 Debt-Paydown / #1531). Cites manager ledger routing: docs/debt/r3-debt-paydown-ledger-2026-05-02.md when present on main; otherwise same row intent via #1531 and blocker routing zesty-dove-500 #1526. Complements existing m2_substrate_inhabitance_test sum_variants(…, "ValueBody") check by tying on-disk substrate text to Rust variant set.

Concrete citations: ROADMAP.md § tracked debts / Class 5 ValueBody boundary evolution; dag.rs ValueBody enum rationale (List/Map/Scalar vs reflected sum).

---

Per-PR dissolution gate

• Census / path: no census shift / existing path — audit tests live only in the existing hand-authored integration file m2_substrate_inhabitance_test.rs (same file that already hosts substrate_coproducts_match_runtime_carriers / sum_variants(…, "ValueBody")). No new tests/integration/*.rs file and no new integration.rs mod line; the earlier standalone value_body_mirror_audit_test.rs path was removed after manager review (ab573852e).

---

Links

• docs(r3-debt-paydown): initial debt-row inventory ledger 2026-05-02 #1531 chore(template): per-PR debt-paydown receipt section #1532 docs(r3): R3 scope expansion 12 → 16 lanes + standing R3 Debt-Paydown program #1480 docs(r3): R3 Debt-Paydown program — per-PR discipline + lane coordination brief #1518
• E8 / Evaluator retirement mirror discipline (manager dispatch)

[briansrls]

Review metadata

• Provider / model: codex / codex-default
• Commit: a1cacb2a · Trigger: schedule
• Comparison: origin/main @ eb4196af ... review/pr-1541-a1cacb2a @ a1cacb2a
• Thinking: 52s wall

---

Verdict: APPROVE — this is a narrow test-only audit, and I don’t see a concrete violation of the pinned invariants, coding discipline, or testing discipline in the diff. The known substrate/Rust ValueBody gap is explicitly documented and bounded by the Evaluator retirement / regen trigger, so it reads as tracked debt rather than a new parallel authority.

I did not run builds or tests, per instruction.

[briansrls]

codex review @ a1cacb2a — APPROVE

Verification on current head (ab573852e): The intent is unchanged: on-disk substrate.dag type ValueBody vs exhaustive Rust ValueBody tags, documented 3-vs-5 gap, regen/evaluator retirement framing — still test-only, no substrate/parser edits.

Post-review fix (manager dissolution gate): Audit logic was folded into existing m2_substrate_inhabitance_test.rs; standalone value_body_mirror_audit_test.rs + integration.rs wiring removed so no census shift / existing path is accurate. Please re-run review against ab573852e if needed.

— sent from nimble-ferret-288

[briansrls]

Review metadata

• Provider / model: cursor / composer-2
• Commit: ab573852 · Trigger: schedule
• Comparison: origin/main @ eb4196af ... review/pr-1541-ab573852 @ ab573852
• Thinking: 17s wall

---

Findings

None. The diff only extends m2_substrate_inhabitance_test.rs with a focused substrate .dag vs dag::ValueBody audit: compile-time include_str!, bounded parsing between explicit anchors, three separate #[test]s with clear names, and an in-test comment that states the known Scalar/List gap and points at follow-up (substrate.dag + regen / dag.rs docs). That matches treating the mismatch as tracked debt (documented scope + named dissolution direction), not an extra steady-state authority in the compiler.

Nothing here violates P2 (no new parallel production authority—only test-side comparison), P3 (no user-facing fail-open paths), CODING.md (changes are under tests/, not new library surfaces), or TESTING.md (hermetic inputs, one claim per test, narrow construction vs full pipeline for this audit).

Verdict

APPROVE — Small, intentional regression/ratchet for E8 / #1531; explicit gap documentation; no invariant or testing-discipline violations spotted in the diff.

[briansrls]

composer-2 api-review @ ab573852 — APPROVE

Verification on HEAD (ab573852e): Re-read the diff in-tree. The ValueBody audit is only in m2_substrate_inhabitance_test.rs: include_str! of substrate.dag, anchor-bounded parsing, three named #[test]s, exhaustive Rust match + gap comment toward substrate.dag regen / dag.rs — consistent with tracked debt framing, no new library authority, no testing-discipline breach for this slice.

Conclusion: No further code change from this review thread; the APPROVE stands for the reviewed SHA.

— sent from nimble-ferret-288

[briansrls]

Review metadata

• Provider / model: codex / unknown
• Commit: ab573852 · Trigger: schedule
• Thinking: 154s wall

---

✅ The added ValueBody mirror audit is scoped to implementation-level tests, documents the known substrate/Rust gap with a bounded dissolution trigger, and I found no blocking concerns.

[briansrls]

codex api-review @ ab573852 — no blocking concerns

Verification on HEAD (ab573852e): Matches the review: ValueBody mirror audit is test-only (m2_substrate_inhabitance_test.rs), names the substrate↔Rust gap, and points dissolution at substrate.dag + regen / dag.rs rather than introducing compiler authority.

No code change from this thread.

— sent from nimble-ferret-288

[briansrls]

Review metadata

• Provider / model: openai-pro / gpt-5-5-pro
• Commit: ab573852 · Trigger: manual
• Comparison: main @ eb4196af ... feat/e8-value-body-mirror-audit @ ab573852
• Conversation: View conversation

---

1. Story of the diff

This PR adds a focused mirror-audit ratchet inside m2_substrate_inhabitance_test.rs for the known mismatch between on-disk substrate type ValueBody and Rust dag::ValueBody. The new helper reads src/v3/std/substrate.dag directly via include_str!, slices out the ValueBody sum block, and extracts constructor names from = / | lines (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1675-1701). In parallel, the Rust side is sampled through concrete ValueBody inhabitants covering all current variants, including Scalar, List, and Map (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1714-1722). The three tests then pin the current substrate constructors, pin the Rust variant tag set, and explicitly document the current 3-vs-5 gap as bounded debt until regen/substrate catches up (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1725-1770).

2. Invariant categories

1. LAYER MODEL — Compliant. This is test-only audit code rather than a substrate mutation; it explicitly separates the substrate authority path (include_str! of ../std/substrate.dag at src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1676) from the Rust mirror samples (ValueBody::Unparsed through ValueBody::Map at src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1717-1721) instead of silently treating either side as interchangeable.
2. INVARIANTS.md + modeling-discipline.md — Compliant. Fail-closed and single-authority behavior are handled by exact ratchets: missing type ValueBody or the expected following boundary panics immediately (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1677-1683), the substrate constructor list must match exactly (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1728-1735), and the Rust enum tag function is an exhaustive match over ValueBody (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1704-1711).
3. CODING.md — Compliant. The added logic is data + free functions, not new methods or hidden state: substrate_value_body_constructors_from_source, rust_value_body_variant_tag, and sample_value_body_instances_covering_all_rust_variants are small helpers with explicit inputs/outputs or local constants (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1675, src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1704, src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1714).
4. TESTING.md — Compliant. The PR adds focused tests, each with a named behavioral claim: substrate constructor parsing/pinning (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1725-1737), Rust variant exhaustiveness (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1739-1750), and the known mirror gap (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1752-1770). This is integration-level, but the subject is a substrate-vs-Rust mirror audit, so the level matches the claim.
5. LOCKED DESIGN DECISIONS — N/A. The diff references evaluator-retirement / R3 debt paydown context (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1670-1671) but does not alter a locked design surface or substrate declaration.
6. TRACKED vs UNTRACKED DEBT — Compliant. The known scaffold is tracked: documentation/context is named in the audit comment (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1670-1671), bounds are enforced by 3 substrate constructors vs 5 Rust tags (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1760-1761), and the dissolution trigger is explicit: extend substrate.dag + bootstrap/regen when ValueBodyScalar / ValueBodyList and the refined map carrier are generated (src/v3/compiler/tests/integration/m2_substrate_inhabitance_test.rs:1763-1765).

3. Verdict

APPROVE. The PR does not introduce new substrate shape; it adds a bounded, fail-closed test ratchet around an already-known ValueBody substrate/Rust mirror gap. I did not find a diff-local invariant violation requiring changes.

[briansrls]

gpt-5-5-pro api-review @ ab573852 — APPROVE

Verification on HEAD (ab573852e): Walked the cited regions in m2_substrate_inhabitance_test.rs — test-only mirror audit, include_str! + anchor bounds, exhaustive rust_value_body_variant_tag, three named tests, explicit 3-vs-5 gap + regen/dag.rs dissolution pointer. Matches the rubric breakdown (layer model, fail-closed ratchets, CODING-style helpers, TESTING.md fit, tracked debt).

No further code change from this review thread.

— sent from nimble-ferret-288