Record D004-PRE-01 reviewed protocol

[chasebryan]

What changed

• Records the owner's acceptance of D004-PRE-01 at the exact immutable review-subject revision.
• Adds the v0.6 reviewed-but-not-executable protocol, canonical 75-record replay plan, and provisional implementation-closure record.
• Adds fail-closed Rust and Python validation for canonical identities, fixture sufficiency, symmetric candidate mappings, deterministic replay, isolation, and result-contract nonclaims.
• Updates architecture, assurance, roadmap, research, traceability, threat-model, compiler, policy, and decision documentation.
• Reseals the downstream D-009 and D-010 semantic bindings and raises repository policy to 0.2.122.

Why

D004-PRE-01 had explicit owner acceptance but still needed a byte-materialized, policy-protected technical disposition before Orange could safely implement execution prerequisites. This tranche closes the seven protocol-review gaps without inventing execution evidence or accepting a semantic-strata candidate.

Scope and impact

D-004 remains proposed. The execution epoch remains null and unfrozen, execution remains unauthorized, evidence remains 0/25 candidate-case units and 0/75 result records, S3 remains incomplete, and S3b remains blocked. The 3-of-10 (30%) gate-closure score is unchanged and is not represented as release readiness. Implementation closure remains provisional until a follow-up record binds these validated bytes to the exact merged revision.

Validation

• make check — passed on the exact commit candidate
• Foundation policy 0.2.122 — passed with 190 required paths and 153 protected files
• Python foundation/decision suites — 412 passed
• D-004 focused Python suite — 13 passed
• D-004/D-009/D-010 focused Python suites — 44 passed
• D-004 Rust suite — 25 passed in debug and release coverage
• Full Rust unit, integration, conformance, documentation, Clippy, formatting, installation, and reproducibility checks — passed
• Independent read-only integration audit — no material findings

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 2c38bb0e2a

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Bind repetition closure to distinct execution identities

When populated result parsing is added, this check can count one captured execution three times: RepetitionRecordSummary contains no execution ordinal or scheduled-execution identity, and the loop only requires callers to relabel otherwise identical summaries as repetitions 1, 2, and 3. Such clones pass validate_repetition_closure, despite the protocol requiring three independent executions and 75 distinct result records. Require each summary to carry and match the expected identity-bound physical schedule record before granting slot closure.

Useful? React with 👍 / 👎.