docs(x10r,D-002H): preregister ricci_flow-scoped canonical run#683
Merged
Conversation
D-002H opens a fresh pre-registered lineage after D-002G structural closure (PR #682 merge 8cf5364). D-002H is scoped to ricci_flow substrate only; block_structured + temporal_coupling are explicitly excluded per the closure finding that they are seed-deterministic by construction (research/systemic_risk/d002c_substrates.py:401). This PR is documentation-only: - D002H_PREREGISTRATION.yaml (locked at this merge commit) - D002H_SCOPE_RATIONALE.md (8 sections) - D002H_CLAIM_BOUNDARY.md (verbatim block) - D002H_CANONICAL_RUN_AUTHORIZATION_GATES.md (7 gates A..G) - artifacts/d002h/prereg/d002h_preregistration_lock.json (schema v1) - tests/systemic_risk/test_d002h_preregistration.py (12 tests) - D002G_CANONICAL_RUN_BLOCKERS.md (append-only lineage-transition note) Non-binding scope statement: D-002H does NOT amend D-002G, does NOT rescue D-002G, does NOT update D002C_CLAIM_LEDGER.yaml, does NOT authorise canonical run, does NOT modify any source code. D-002C ledger sha256 f96ba9b5a2057d2e0bff84afc28578ab316cff73f6dc6673fb0d6d543b8bd6dd verified byte-exact unchanged. Substrate code byte-exact unchanged. Canonical D-002H run requires the explicit conjunction A AND B AND C AND D AND E AND F AND G per D002H_CANONICAL_RUN_AUTHORIZATION_GATES.md. Each gate is a separate downstream artifact; THIS PR passes Gate A only. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 7eba41cde4
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".
| included list is forbidden. | ||
| - **Any aggregated claim across substrate types.** The aggregate | ||
| estimator denominator drops from 3 substrates to 1; statistical | ||
| power loss is √3 ≈ 1.7× standard-error reduction for any aggregated |
There was a problem hiding this comment.
This sentence states that narrowing from 3 substrates to 1 yields a 1.7× standard-error reduction, but with fewer independent substrate rows the standard error scales in the opposite direction (it increases by about √3). Keeping it as written inverts the statistical implication of the scope change and can mislead later interpretation of power/uncertainty tradeoffs in this governance record.
Useful? React with 👍 / 👎.
…detect-secrets baseline CI fail on PR #683 secrets-supply-chain: 3 sha256 hex hits across - artifacts/d002h/prereg/d002h_preregistration_lock.json:6 (parent_merge_sha) - docs/governance/D002H_PREREGISTRATION.yaml:30 (parent_merge_sha) - tests/systemic_risk/test_d002h_preregistration.py:50 (D-002G closure sha pin) All three are content-anchors over D-002G structural closure merge commit 8cf5364 — NOT credentials. Same canonical fix as PRs #681 / #682: regenerate .github/detect-secrets.baseline to register the 3 new hits as known-false-positives. Baseline grows 95 → 98 tracked files (3 new file entries added). Local verification: detect-secrets-hook --baseline .github/detect-secrets.baseline \ artifacts/d002h/prereg/d002h_preregistration_lock.json \ docs/governance/D002H_PREREGISTRATION.yaml \ tests/systemic_risk/test_d002h_preregistration.py → exit 0 pytest tests/systemic_risk/test_d002h_preregistration.py → 12 passed No prereg edit. No code change. No test change. No governance touch. D002C ledger sha256 f96ba9b5a2057d2e0bff84afc28578ab316cff73f6dc6673fb0d6d543b8bd6dd unchanged. Substrate code unchanged. Mechanism code unchanged. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
7 tasks
neuron7xLab
pushed a commit
that referenced
this pull request
May 13, 2026
… invariant naming The Gate B agent landed 10 tests covering the necessary invariants under my naming convention. The operator's contract requires 8 additional explicit test names so the per-rung invariant naming matches the brief verbatim (and so future audits can grep for the exact contract names). Added 8 tests with the contract names: - test_gate_b_scope_ricci_flow_only - test_gate_b_reverifies_m1 - test_gate_b_reverifies_m3 - test_gate_b_forbids_block_structured - test_gate_b_forbids_temporal_coupling - test_gate_b_does_not_authorize_canonical_run - test_gate_b_preserves_prereg_lock - test_gate_b_no_cross_substrate_claim Each new test asserts the contract invariant against the existing Gate B artifact (artifacts/d002h/eligibility/d002h_ricci_eligibility.json) or the Gate B report (docs/governance/D002H_GATE_B_REPORT.md). No artifact recomputation; no script change; no governance edit; no mechanism code touch. test_gate_b_preserves_prereg_lock pins the D-002H pre-registration sha256 at the PR #683 anchor (44b18b5a40ce9d188a9c3bd49339621f81a65a15f97a683247902450dd54acec) with inline # fmt: off + # noqa: E501 + # pragma: allowlist secret per the established pattern. test_gate_b_no_cross_substrate_claim scans the Gate B report for cross-substrate phrases outside denial context. Total Gate B tests now: 18 (10 prior + 8 contract aliases). All pass. Black + ruff format + ruff lint + mypy --strict clean. No change to artifact, script, or report. No mechanism / substrate / ledger / prereg touch. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
neuron7xLab
pushed a commit
that referenced
this pull request
May 13, 2026
…etect-secrets baseline The 4be266d commit added user-contract test aliases including D002H_PREREG_SHA256 = "44b18b5a40ce9d188a9c3bd49339621f81a65a15f97a683247902450dd54acec" as an inline content-anchor over the D-002H pre-registration at the PR #683 merge. The inline `# pragma: allowlist secret` covers the literal, but detect-secrets-hook in baseline-validation mode also requires the baseline file to register the line. Same canonical fix as PR #681/#682/#683 baseline updates: regenerate .github/detect-secrets.baseline to include the new line as a known-false-positive. No code change. No test change. No governance touch. D002C ledger sha256 f96ba9b5a2057d2e0bff84afc28578ab316cff73f6dc6673fb0d6d543b8bd6dd unchanged. D-002H prereg sha unchanged. Mechanism + substrate code unchanged. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
neuron7xLab
pushed a commit
that referenced
this pull request
May 13, 2026
…t codebase state Hook on CI was returning exit code 3 (baseline auto-mutated, commit required) because the pre-existing baseline's per-line entries were stale relative to the current branch's file states. The inline pragmas on the test file's two sha256 anchors (D002C_LEDGER_SHA256 + D002H_PREREG_SHA256) ARE respected by both `detect-secrets scan` and `detect-secrets-hook` — confirmed by exit-0 on both invocations post-regen. Canonical fix (same pattern as PR #683 ae3f8b3): regenerate baseline to reflect current codebase. The regen drops orphan entries for files whose hex strings no longer trigger on the current branch (105 → 101 files). The 4 dropped entries had no current matches; they are preserved in git history and will be re-added if the underlying files ever regress. Local verification: detect-secrets scan --baseline ... → regenerated detect-secrets-hook --baseline ... <changed-files> → exit 0 pytest tests/systemic_risk/test_d002h_gate_b_eligibility.py → 18 passed No code change. No test change. No governance touch. D002C ledger sha256 unchanged. D-002H prereg sha unchanged. Mechanism + substrate code unchanged. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
neuron7xLab
added a commit
that referenced
this pull request
May 13, 2026
… artifact (#684) * docs(x10r,D-002H,gate-B): ricci_flow M1/M3 eligibility reverification artifact D-002H Gate B per D002H_CANONICAL_RUN_AUTHORIZATION_GATES.md §B: - run verify_m1_eligibility + verify_m3_eligibility on ricci_flow × {N=50,100,200} × λ ∈ {0.0, 0.05, 0.10, 0.20, 0.40, 1.0} (18 cells) - emit machine-readable verdict artifact at artifacts/d002h/eligibility/d002h_ricci_eligibility.json (schema v1) - 10-test scaffolding verifying schema + invariants + D002C ledger byte-exact unchanged - BLOCKERS.md append-only Gate B status section Reuses existing verifiers from research/systemic_risk/d002g_null_mechanisms.py (sha-pinned at the prior D-002G merge); no new mechanism, no new verdict literals, no new salt, no substrate/mechanism code edit. λ=0 cells emit N/A_M3_REQUIRES_LAMBDA_GT_ZERO per the M3 module's contract. M1 admissibility (no public verify_m1_eligibility symbol exists) is evaluated by invoking realize_null(strategy="M1_INDEPENDENT_SEED", ...) and translating its outcome (success / BitIdenticalNullError / unexpected) into ELIGIBLE_M1 / INELIGIBLE_M1_BIT_IDENTICAL / INDETERMINATE_M1_PROVENANCE_MISSING — no new verdict literal coined. Verdict: PASS (18/18 cells). At every λ>0 cell the M3 verifier returns ELIGIBLE_M3 with the marginal-match report inside the locked M3 tolerances; M1 returns ELIGIBLE_M1 on every cell including λ=0. Gate B PASS is necessary but NOT sufficient for canonical run authorisation. Conjunction A ∧ B ∧ C ∧ D ∧ E ∧ F ∧ G is the contract; this PR addresses Gate B alone. Gates C–G remain open. Canonical run remains BLOCKED. D002C ledger sha256 f96ba9b5a2057d2e0bff84afc28578ab316cff73f6dc6673fb0d6d543b8bd6dd verified byte-exact unchanged. D-002H prereg sha256 44b18b5a40ce9d188a9c3bd49339621f81a65a15f97a683247902450dd54acec verified byte-exact unchanged at PR #683 anchor. detect-secrets baseline updated to register 35 known-false-positive sha256 hex hits across the new artifact + compute-script. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * test(x10r,D-002H,gate-B): add user-contract test aliases for explicit invariant naming The Gate B agent landed 10 tests covering the necessary invariants under my naming convention. The operator's contract requires 8 additional explicit test names so the per-rung invariant naming matches the brief verbatim (and so future audits can grep for the exact contract names). Added 8 tests with the contract names: - test_gate_b_scope_ricci_flow_only - test_gate_b_reverifies_m1 - test_gate_b_reverifies_m3 - test_gate_b_forbids_block_structured - test_gate_b_forbids_temporal_coupling - test_gate_b_does_not_authorize_canonical_run - test_gate_b_preserves_prereg_lock - test_gate_b_no_cross_substrate_claim Each new test asserts the contract invariant against the existing Gate B artifact (artifacts/d002h/eligibility/d002h_ricci_eligibility.json) or the Gate B report (docs/governance/D002H_GATE_B_REPORT.md). No artifact recomputation; no script change; no governance edit; no mechanism code touch. test_gate_b_preserves_prereg_lock pins the D-002H pre-registration sha256 at the PR #683 anchor (44b18b5a40ce9d188a9c3bd49339621f81a65a15f97a683247902450dd54acec) with inline # fmt: off + # noqa: E501 + # pragma: allowlist secret per the established pattern. test_gate_b_no_cross_substrate_claim scans the Gate B report for cross-substrate phrases outside denial context. Total Gate B tests now: 18 (10 prior + 8 contract aliases). All pass. Black + ruff format + ruff lint + mypy --strict clean. No change to artifact, script, or report. No mechanism / substrate / ledger / prereg touch. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(x10r,D-002H,gate-B): register new D002H_PREREG_SHA256 anchor in detect-secrets baseline The 4be266d commit added user-contract test aliases including D002H_PREREG_SHA256 = "44b18b5a40ce9d188a9c3bd49339621f81a65a15f97a683247902450dd54acec" as an inline content-anchor over the D-002H pre-registration at the PR #683 merge. The inline `# pragma: allowlist secret` covers the literal, but detect-secrets-hook in baseline-validation mode also requires the baseline file to register the line. Same canonical fix as PR #681/#682/#683 baseline updates: regenerate .github/detect-secrets.baseline to include the new line as a known-false-positive. No code change. No test change. No governance touch. D002C ledger sha256 f96ba9b5a2057d2e0bff84afc28578ab316cff73f6dc6673fb0d6d543b8bd6dd unchanged. D-002H prereg sha unchanged. Mechanism + substrate code unchanged. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * Revert "fix(x10r,D-002H,gate-B): register new D002H_PREREG_SHA256 anchor in detect-secrets baseline" This reverts commit b54a3ea. * fix(x10r,D-002H,gate-B): correct stale .bak filter reference in detect-secrets baseline Prior baseline regenerations left a stale reference to `.github/detect-secrets.baseline.bak` in the `filename:` field of the `is_baseline_file` filter (line 92). This caused the filter to NOT match the actual baseline path, leaving the new D002H_PREREG_SHA256 inline pragma in test_d002h_gate_b_eligibility.py unrecognised by the hook. Two-line fix: - filename: ".github/detect-secrets.baseline.bak" → ".github/detect-secrets.baseline" - generated_at: timestamp bump After this fix, detect-secrets-hook exits 0 on the test file without adding a new baseline entry — the inline `# pragma: allowlist secret` is now correctly recognised once the filter targets the real path. No baseline file count change (105 files preserved). No test change. No governance touch. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(x10r,D-002H,gate-B): regenerate detect-secrets baseline to current codebase state Hook on CI was returning exit code 3 (baseline auto-mutated, commit required) because the pre-existing baseline's per-line entries were stale relative to the current branch's file states. The inline pragmas on the test file's two sha256 anchors (D002C_LEDGER_SHA256 + D002H_PREREG_SHA256) ARE respected by both `detect-secrets scan` and `detect-secrets-hook` — confirmed by exit-0 on both invocations post-regen. Canonical fix (same pattern as PR #683 ae3f8b3): regenerate baseline to reflect current codebase. The regen drops orphan entries for files whose hex strings no longer trigger on the current branch (105 → 101 files). The 4 dropped entries had no current matches; they are preserved in git history and will be re-added if the underlying files ever regress. Local verification: detect-secrets scan --baseline ... → regenerated detect-secrets-hook --baseline ... <changed-files> → exit 0 pytest tests/systemic_risk/test_d002h_gate_b_eligibility.py → 18 passed No code change. No test change. No governance touch. D002C ledger sha256 unchanged. D-002H prereg sha unchanged. Mechanism + substrate code unchanged. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(x10r,D-002H,gate-B): fix required_python_symbols schema in commit acceptor CommitAcceptor pydantic schema expects required_python_symbols as a list of dotted-path STRINGS (e.g. "research.systemic_risk.module.symbol"), matching the format used by all prior D-002G acceptors (e.g. x10r-d002g-p3-constant-payload-null-recovery.yaml). The Gate B agent produced a list of object literals {module, symbol}, which trips 4 string_type validation errors in tests/governance/test_typed_models.py::test_canonical_acceptor_corpus_parses. Fix: flatten the 4 entries to dotted-path strings. No semantic change — the symbols and the modules they live in are unchanged; only the representation is corrected. Local verification: pytest tests/governance/test_typed_models.py::test_canonical_acceptor_corpus_parses → PASS No code change. No test change. No governance touch. No baseline change. D002C ledger sha unchanged. D-002H prereg sha unchanged. Mechanism + substrate code unchanged. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Yaroslav Vasylenko <neuron7x@ukr.net> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
8 tasks
neuron7xLab
added a commit
that referenced
this pull request
May 13, 2026
Gate C of the locked 7-gate D-002H authorisation conjunction (A AND B AND C AND D AND E AND F AND G) closes here as a pure declaration: the canonical parameter grid is pinned, byte-equivalent to the ``canonical_grid`` block of the locked D-002H pre-registration (sha256 44b18b5a...0dd54acec, PR #683 anchor), as a content-addressed artifact at artifacts/d002h/canonical/d002h_canonical_grid.json under schema D002H-CANONICAL-GRID-v1. * substrates: [ricci_flow] (exclusions per prereg) * N: [50, 100, 200] * lambda: [0.0, 0.05, 0.10, 0.20, 0.40, 1.0] * total: 18 cells (3 N x 6 lambda) * n_seeds: 20 * n_bootstrap:16 * base_seed: 42 (matches Gate B base_seed) * null_seed_M3: 12345 Gate C does NOT authorise canonical run. It does NOT execute any sweep. It does NOT produce results. canonical_run_authorized stays false; downstream_gates_remaining == ["D","E","F","G"]. The 7-gate conjunction is the canonical-run authorisation contract; this PR addresses Gate C alone. Locked-file invariants preserved (byte-exact): * D-002H prereg sha256 unchanged at PR #683 anchor. * D-002C claim ledger sha256 unchanged at the canonical anchor. * substrate + null-mechanism module shas unchanged. * No edit to any prior gate artifact or report. * No new mechanism, no new verdict literal, no new salt. Files (5): * artifacts/d002h/canonical/d002h_canonical_grid.json (new, schema v1) * docs/governance/D002H_GATE_C_CANONICAL_GRID.md (new report) * tests/systemic_risk/test_d002h_gate_c_grid.py (10 invariant tests) * .claude/commit_acceptors/x10r-d002h-gate-c-canonical-grid.yaml * docs/governance/D002G_CANONICAL_RUN_BLOCKERS.md (append-only) Plus .github/detect-secrets.baseline regenerated to cover the new grid JSON + report hex anchors (governance hashes, not credentials). Quality gates (local, scoped): * pytest gate-C: 10/10 PASS * pytest typed_models: 12/12 PASS (acceptor parses Pydantic v2) * pytest d002h+d002g k-filter: all PASS * ruff format --check / ruff check: clean * black --check: clean * mypy --strict --follow-imports=silent: clean * detect-secrets-hook: exit 0 Co-authored-by: Yaroslav Vasylenko <neuron7x@ukr.net> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
neuron7xLab
pushed a commit
that referenced
this pull request
May 13, 2026
…baseline CI fail on PR #687 secrets-supply-chain: artifacts/d002h/locks/ d002h_locked_file_pins.json contains 16 content-anchor sha256 entries (one per locked file pinned at Gate D anchor 077073e). JSON does not support inline pragma comments — the agent's L2 application covered only the test file's 2 inline anchors, missing the JSON's 16. Same canonical fix as PRs #681/#682/#683/#684 baseline updates: regenerate .github/detect-secrets.baseline to register all 16 JSON hits as known-false-positives. Local verification: detect-secrets-hook --baseline ... d002h_locked_file_pins.json → exit 0 pytest tests/systemic_risk/test_d002h_gate_e_locked_files.py → 25 passed No code change. No test change. No governance touch. No prereg edit. No source code edit. All 16 pinned shas remain byte-exact at Gate D anchor: D002C ledger f96ba9b5...d6dd unchanged, D-002H prereg unchanged, substrate + mechanism code unchanged. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
neuron7xLab
added a commit
that referenced
this pull request
May 13, 2026
* test(x10r,D-002H,gate-E): locked-ledger verification artifact Gate E of the 7-gate D-002H canonical-run authorisation conjunction (A AND B AND C AND D AND E AND F AND G). Pure declaration + byte-exact sha verification: emits a machine-readable artifact at artifacts/d002h/locks/d002h_locked_file_pins.json (schema D002H-GATE-E-v1) and a human-readable report at docs/governance/D002H_GATE_E_LOCKED_FILE_REPORT.md. The artifact pins 16 files at the Gate D merge anchor (077073e, PR #686): - 4 D-002C governance anchors (claim ledger + prereg + canonical-run report + null-audit falsification report). - 6 D-002G governance anchors (prereg + non-degenerate null design + acceptance rules + P3/M3 prereg + structural closure + negative space map). - 4 D-002H governance anchors (prereg + scope rationale + claim boundary + canonical-run authorisation gates doc). - 2 source-code modules (d002c_substrates.py + d002g_null_mechanisms.py). The test suite tests/systemic_risk/test_d002h_gate_e_locked_files.py (25 tests = 10 base contract tests + 15 parametrised pin cases over the 16 pinned files) recomputes sha256 from disk bytes for each pin and asserts byte-exact equality. Any single-byte mutation in any pinned file fails Gate E. Scope (per the operating law): - NO canonical run. NO sweep. NO results. - NO new mechanism. NO new verdict literal. - NO substrate / mechanism code edit. - NO D-002C ledger mutation (sha pinned at f96ba9b5...d6dd). - NO D-002H prereg edit (sha pinned at 44b18b5a...4acec). - NO claim "Gate E PASS -> canonical run authorised"; the 7-gate conjunction A AND B AND C AND D AND E AND F AND G is the contract. D-002G_CANONICAL_RUN_BLOCKERS.md gains an append-only "D-002H Gate E" section after the Gate D section. .github/detect-secrets.baseline: timestamp-only refresh (no new entries; the two inline hex anchors in the test file carry # pragma: allowlist secret per the Gate D acceptor's pattern). Gates A, B, C, D, E closed. Gates F, G remain open. Canonical run remains BLOCKED until A AND B AND C AND D AND E AND F AND G. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(x10r,D-002H,gate-E): register 16 JSON pin shas in detect-secrets baseline CI fail on PR #687 secrets-supply-chain: artifacts/d002h/locks/ d002h_locked_file_pins.json contains 16 content-anchor sha256 entries (one per locked file pinned at Gate D anchor 077073e). JSON does not support inline pragma comments — the agent's L2 application covered only the test file's 2 inline anchors, missing the JSON's 16. Same canonical fix as PRs #681/#682/#683/#684 baseline updates: regenerate .github/detect-secrets.baseline to register all 16 JSON hits as known-false-positives. Local verification: detect-secrets-hook --baseline ... d002h_locked_file_pins.json → exit 0 pytest tests/systemic_risk/test_d002h_gate_e_locked_files.py → 25 passed No code change. No test change. No governance touch. No prereg edit. No source code edit. All 16 pinned shas remain byte-exact at Gate D anchor: D002C ledger f96ba9b5...d6dd unchanged, D-002H prereg unchanged, substrate + mechanism code unchanged. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Yaroslav Vasylenko <neuron7x@ukr.net> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
6 tasks
neuron7xLab
added a commit
that referenced
this pull request
May 13, 2026
…OMPLETE (#689) D-002H Gate G TERMINAL closure of the 7-gate canonical-run authorisation conjunction (A ∧ B ∧ C ∧ D ∧ E ∧ F ∧ G all PASS). Phase 0 verification of prior gate CI: A #683 1b59ce5 — ALL_REQUIRED_PASS at merge B #684 b97daae — ALL_REQUIRED_PASS at merge C #685 a9d852d — ALL_REQUIRED_PASS at merge D #686 077073e — ALL_REQUIRED_PASS at merge E #687 e1d3ae3 — ALL_REQUIRED_PASS at merge F #688 0e598ff — ALL_REQUIRED_PASS at merge This PR emits the TERMINAL authorisation state: status: CANONICAL_RUN_AUTHORIZED canonical_run_authorized_final: true authorisation_scope: ricci_flow substrate only The canonical sweep itself is a SEPARATE downstream PR; Gate G does NOT execute the sweep. D-002C ledger byte-exact unchanged (f96ba9b5...d6dd). D-002H prereg byte-exact unchanged (44b18b5a...acec). No substrate / mechanism / governance source touch. Co-authored-by: Yaroslav Vasylenko <neuron7x@ukr.net> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Promise
D-002H opens a fresh pre-registered lineage scoped to
ricci_flowonly, following the D-002G structural closure (PR #682 merge8cf5364a). THIS PR is documentation-only and PASSES Gate A only of the 7-gate canonical-run authorisation conjunction.Lineage relationship to D-002G
D-002H is not an amendment of D-002G. D-002H is not a rescue of D-002G or D-002C. The 4 negative-result PRs (#677 / #679 / #680 / #681) plus the structural closure (#682) remain sha-pinned as the append-only canonical record. D-002H opens a fresh lineage that explicitly acknowledges substrate exclusion and locks a narrowed canonical grid.
origin/main):8cf5364a3f3b605d8b134bccbfe5170098e0e197docs/governance/D002G_STRUCTURAL_CLOSURE_REPORT.mddocs/governance/D002G_CANONICAL_RUN_BLOCKERS.md(append-only)Scope
substrate_scope.included[ricci_flow]substrate_scope.excluded[block_structured, temporal_coupling]null_mechanisms_allowed[M1_INDEPENDENT_SEED, M3_TOPOLOGY_CONDITIONED]canonical_grid.total_cellscanonical_run_authorizedfalserequires_explicit_authorization_artifacttruelineage_typefresh_preregistrationprereg_lock.edit_policyfresh_pre_registration_onlyExclusion root cause:
research/systemic_risk/d002c_substrates.py:401—BlockStructuredSubstrate.realize()discards seed by construction;TemporalKtSubstrate.realize()inherits via delegation at lines 481-483.Allowed claims
Forbidden claims (extract)
7 authorisation gates summary
artifacts/d002h/eligibility/d002h_ricci_eligibility.json(downstream).artifacts/d002h/canonical/d002h_canonical_grid.json(downstream).artifacts/d002h/authorization/d002h_canonical_run_authorisation.json(separate downstream PR).Conjunction: canonical run allowed iff A ∧ B ∧ C ∧ D ∧ E ∧ F ∧ G. THIS PR passes Gate A only. Gates B–G live in separate downstream artifacts.
Tests
tests/systemic_risk/test_d002h_preregistration.py(12 tests, all PASS locally):test_d002h_prereg_existstest_d002h_scope_is_ricci_flow_onlytest_d002h_excludes_block_structuredtest_d002h_excludes_temporal_couplingtest_d002h_forbids_cross_substrate_claimstest_d002h_forbids_d002g_rescue_claimtest_d002h_forbids_canonical_run_without_authorizationtest_d002h_preserves_d002c_ledger(pins D-002C ledger sha tof96ba9b5...d6dd)test_d002h_parent_closure_sha_pinnedtest_d002h_requires_fresh_lineagetest_d002h_claim_boundary_verbatim_presenttest_d002h_no_substrate_code_touch(pinsd002c_substrates.pysha to base-anchor4b2e5d65...0eca)Claim boundary (verbatim, from
D002H_CLAIM_BOUNDARY.md)Forbidden interpretations (from
D002H_CLAIM_BOUNDARY.md)Test plan
ruff format --checkclean on test fileruff checkclean on test fileblack --checkclean on test filemypy --strict --follow-imports=silentclean on test filef96ba9b5...d6dd🤖 Generated with Claude Code