# Non-Core Evidence Ledger

Last updated: 2026-07-17

This ledger makes the currently accepted non-core evidence transitions visible
without promoting any chapter core claim. It is a public trust surface for
readers, reviewers, and future writing agents.

## Current Boundary

| Field | Status |
|---|---|
| Chapter core claims | All 87 remain at `argument`. |
| Accepted non-core upward transitions | 25 narrow transitions. |
| Accepted historical `refuted` labels | 3 immutable transition labels: two QCSA fixtures and the frozen KERC proxy result. Retrospective competence classification is 2 N2 and 1 N1, respectively; none earns exact, mechanism, architecture, parent, or chapter-core refutation. |
| Accepted terminal full-attempt dispositions | 2 post-v2.3 no-promotion decisions: routing/deliberation mixed and update/unlearning claim narrowed. |
| Accepted no-promotion side-lane decisions | 61 accepted `blocks_promotion` decisions; no support-state movement. |
| Accepted terminal instrument-failure dispositions | 1 three-version Campaign 4 repair; heldout sealed and no claim attempt. |
| Accepted live claim-surface narrowing records | 1 count-surface correction; no support-state movement. |
| Chapter-core promotion effect | None. |
| Canonical identity resolution | All 115 accepted transitions resolve through 25 exact atom, 61 bounded subclaim, and 29 proxy relations in `evidence_quality/claim_identity_graph.json`; indirect relations have no parent support-state effect. |
| Negative-result competence | All 90 accepted negative/no-change transitions retain their raw outcomes and are classified as 1 N0, 15 N1, 74 N2, and 0 N3–N5 in `evidence_quality/negative_result_rehabilitation.json`; identity mapping is not competence evidence. |
| External review status | No independent external-human review was requested, required, or claimed for this internal adjudication. |
| No-promotion side-lane record coverage | The 61 accepted records cover the prior 40 v1.x side lanes, three post-v2 programs, six post-v2.1 result claims, five QCSA implementation/evaluation boundaries, and seven post-v2.3 campaign or terminal-batch dispositions, including the repaired flagship, sealed Campaign 4 instrument failure, routing/deliberation mixed result, update/unlearning axis narrowing, and rollback-outcome narrowing. They are blocking decisions, not upward support-state transitions. None promotes a chapter core, deployment, benchmark, model-quality, safety, security, AGI, or ASI claim. |

## Accepted Non-Core Transitions

| Claim ID | New support state | Evidence packet | What moved | What did not move |
|---|---|---|---|---|
| `circle-calculus-and-proof-carrying-ai-contracts.mechanism.003` | `prototype-backed` | `docs/p5_mandatory_terminal_batch.md`; `evidence_transitions/post_v2_3/p5_circle_named_target_receipt_prototype_backed.json` | One pinned Circle target compiled with retained command/result receipts, an explicit proof boundary, and four rejecting consumer controls. | No chapter core, general proof transport, semantic refinement, multi-consumer benefit, transfer, SOTA, AGI, or ASI claim moved. |
| `system-boundaries-and-authority.invariant.001` | `synthetic-test-backed` | `docs/p5_mandatory_terminal_batch.md`; `evidence_transitions/post_v2_3/p5_authority_no_silent_expansion_synthetic_test_backed.json` | One finite sequential grant-to-effect model retained two pre-effect denials, zero governed unsafe releases, and 38/38 rejected authority mutations. | No chapter core, authenticated identity, deployed enforcement, complete mediation, distributed revocation, open-world safety, SOTA, AGI, or ASI claim moved. |
| `kerc.protected_exact_handle_preservation` | `synthetic-test-backed` | `docs/p4_kerc_runtime_campaign.md`; `evidence_transitions/post_v2_3/kerc_protected_exact_handle_synthetic_test_backed.json` | On 64 authored bilingual held-out records, protected-object recall and lossless recovery were 1.000000; removing protected-span capture reduced object recall to zero. | No chapter core, general semantics, multilingual transfer, privacy, security, production efficiency, SOTA, AGI, or ASI claim moved. |
| `kerc.interaction_shared_glossary_break_even` | `synthetic-test-backed` | `docs/p4_kerc_runtime_campaign.md`; `evidence_transitions/post_v2_3/kerc_shared_glossary_break_even_synthetic_test_backed.json` | One frozen repeated-terminology byte curve crossed the simple glossary-handle break-even at two turns. | No end-to-end compression, semantic, privacy, migration, model-compute, energy, governance, production, SOTA, AGI, or ASI claim moved. |
| `situated-world-model.finite-pomdp-governed-acquisition-and-consolidation` | `synthetic-test-backed` | `docs/p4_situated_world_model_campaign.md`; `evidence_transitions/post_v2_3/situated_world_model_finite_pomdp_synthetic_test_backed.json` | One prospectively frozen 11,250-episode, two-environment, five-seed, ten-arm campaign passed all integrity and baseline gates plus six aggregate causal-ablation signatures; the governed path reached 0.7933/0.7967 hidden-state accuracy, 0.1000/0.1133 unsafe-action rates, and 10/10 exact governed replacement and rollback receipts. | No chapter core, neural-world-model, learned-representation, natural-task, open-world, general-causality, reliable-shift-detection, simulator-validity, sim-to-real, deployment, SOTA, AGI, or ASI claim moved; active information was heterogeneous and task success traded against escalation. |
| `governed-usefulness.held-out-local-policy-effect` | `synthetic-test-backed` | `docs/p4_governed_usefulness_campaign.md`; `evidence_transitions/post_v2_3/governed_usefulness_held_out_local_promote.json` | On one prospectively frozen 16-task held-out local corpus and one quantized Qwen3-8B run, full governance released 9 useful and 0 unsafe candidates versus 0 useful and 0 unsafe releases under candidate self-release; removing evidence freshness exposed one additional useful-unsafe release. | No chapter-core claim, model/workload transfer, deployed policy, open-world safety, effect-complete rollback, independent external validity, acceptable governance-cost, SOTA, AGI, or ASI claim moved. |
| `living-book-methodology.phase5_harness_registry_runner` | `synthetic-test-backed` | `docs/phase5_harness_runner.md`; `evidence_transitions/v1_0_measured/phase5_harness_runner_synthetic_test_backed.json` | The repository-infrastructure claim that the Phase 5 harness registry can be replayed by one local runner command, executes all registered synthetic harnesses, and matches each registry result summary. | No chapter core claim, deployed runtime behavior, model quality, benchmark quality, source interpretation, safety, alignment, or governance-effectiveness claim moved. |
| `resource-economics.costed_route_budget_slice` | `synthetic-test-backed` | `docs/costed_route_resource_slice.md`; `evidence_transitions/v1_0_measured/costed_route_resource_slice_synthetic_test_backed.json` | The bounded synthetic costed-route/resource-budget selector slice that rejects a cheap failed negative control, keeps an adequate overkill baseline, and selects the lowest-cost eligible route in the public-safe fixture. | No chapter core claim, deployed routing, scheduler, runtime, load, KV-cache, economic, benchmark, model-quality, safety, or source-interpretation claim moved. |
| `resource-economics.finite_burst_load_smoothing_selector` | `synthetic-test-backed` | `docs/resource_load_stability_probe.md`; `evidence_transitions/v1_x_measured/resource_load_stability_selector_synthetic_test_backed.json` | The bounded finite synthetic load-smoothing selector claim that a 10-task burst-review workload selects protected capacity smoothing over admit-arrivals, reduces instability units from 5 to 0, residualizes 7 deferred task-ticks, and rejects a cheaper review-erasure negative control. | No chapter core claim, real load-stability, deployed scheduler, TokenMana, PlanForge, reviewer-optimization, human-productivity, economic, benchmark, model-quality, safety, or source-interpretation claim moved. |
| `resource-economics.scoped_workflow_trace_route_selector` | `empirical-test-backed` | `docs/resource_workload_quality_probe.md`; `evidence_transitions/v1_x_measured/resource_workload_quality_selector_empirical_test_backed.json` | The bounded local repository-task selector claim that five measured samples per route select the scoped workflow-trace validator over the broader Resource live-probe baseline while rejecting a cheaper no-op success-text command that exits 0 without producing the required validation surface. | No chapter core claim, broader workload-quality claim, stable-speedup claim, deployed scheduler, TokenMana, PlanForge, production workload, economic, benchmark, model-quality, safety, or source-interpretation claim moved. |
| `circle-calculus.external_rope_receipt_replay` | `prototype-backed` | `docs/circle_external_receipt_slice.md`; `evidence_transitions/v1_0_measured/circle_external_rope_receipt_prototype_backed.json` | The bounded imported external-prototype receipt fact that a clean local Circle checkout at commit `63b0f511` built the Circle target, certified one rope position distinguishability contract, emitted/accepted the recorded receipt, and passed the selected receipt/contract test batch summarized in the public-safe result record. | No chapter core claim, deployed proof-contract transport, model quality, reasoning ability, context length, speed, memory scaling, transfer, benchmark, safety, or ASI claim moved. |
| `compact-generative-systems.compact_gvr_receipt_slice` | `synthetic-test-backed` | `docs/compact_gvr_slice.md`; `evidence_transitions/v1_x_measured/compact_gvr_slice_synthetic_test_backed.json` | The bounded synthetic compact-generation/GVR receipt claim that a local validator compares a 368-byte literal baseline to a 78-byte exact compact generator-plus-repair receipt, rejects lossy exactness, negative-rate/no-fallback, and bounded-search-overrun controls, and checks a finite Lean fixture bridge. | No chapter core claim, deployed compression, codec correctness, semantic utility, fallback execution, benchmark, model-quality, safety, ASI, or source-interpretation claim moved. |
| `project-theseus-as-report-first-implementation-reference.artifact_retention_replay_gate_import` | `prototype-backed` | `docs/theseus_artifact_retention_replay_import.md`; `evidence_transitions/v1_x_measured/theseus_artifact_retention_replay_import_prototype_backed.json` | The bounded sanitized Project Theseus artifact-retention replay import records one exact-hash retained-artifact replay, 41,943,527 replayed payload bytes, 2,389,576 archived bytes, one compressed-artifact record, one compression receipt, one proof-contract receipt, one claim record, one artifact-graph record, one storage evidence-transition record, one defeater record, zero public training rows, zero external inference calls, and seven expected-invalid controls. | No Project Theseus chapter core claim, clean live Project Theseus replay, deployed residual-ledger storage, deployed artifact-graph behavior, model quality, benchmark performance, generation speed, safety, alignment, transfer, deployment readiness, or ASI claim moved. |
| `project-theseus-as-report-first-implementation-reference.module_definition_of_done_gate_import` | `prototype-backed` | `docs/theseus_module_definition_of_done_import.md`; `evidence_transitions/v1_x_measured/theseus_module_definition_of_done_import_prototype_backed.json` | The bounded sanitized Project Theseus module definition-of-done gate import records trigger state `GREEN`, 22 of 22 major module records ready, 0 hard gaps, 0 warnings, all seven book-standard sources present, negative evidence linked, 20 source-backlog work cards, eight steward decisions, and seven expected-invalid controls. | No Project Theseus chapter core claim, clean live Project Theseus replay, module capability, deployed Theseus behavior, model quality, benchmark performance, safety, alignment, transfer, deployment readiness, or ASI claim moved. |
| `project-theseus-as-report-first-implementation-reference.project_registry_reality_import` | `prototype-backed` | `docs/theseus_project_registry_import.md`; `evidence_transitions/v1_x_measured/theseus_project_registry_import_prototype_backed.json` | The bounded sanitized Project Theseus project-registry import records trigger state `GREEN`, 5,662 registered paths, 24 owned lifecycle surfaces, full registry coverage, 0 unregistered active sources, 0 stale or missing report outputs, 0 generated source artifacts, 0 registry-governance violations, 0 external inference calls, and nine expected-invalid controls. | No Project Theseus chapter core claim, clean live Project Theseus replay, deployment, model quality, benchmark performance, generation speed, self-evolution safety, alignment, transfer, deployment readiness, or ASI claim moved. |
| `project-theseus-as-report-first-implementation-reference.assistant_reference_trace_import` | `prototype-backed` | `docs/theseus_assistant_reference_trace_import.md`; `evidence_transitions/v1_x_measured/theseus_assistant_reference_trace_import_prototype_backed.json` | The bounded sanitized Project Theseus assistant reference-trace import records source digests, 19 required reference-trace record types, 27/27 gates, 2,203 VIEA view records, 12 selected VCM pages, zero public training rows, zero external inference calls, no copied raw report, raw trace, raw assistant text, raw prompt, or private payload, and 11 expected-invalid controls. | No Project Theseus chapter core claim, clean live Project Theseus replay, current runtime state, deployed behavior, route quality, private verifier quality, model quality, benchmark superiority, useful-solution-per-second improvement, safety, alignment, transfer, deployment readiness, or ASI claim moved. |
| `project-theseus-as-report-first-implementation-reference.accelerator_parity_manifest_import` | `prototype-backed` | `docs/theseus_accelerator_parity_manifest_import.md`; `evidence_transitions/v1_x_measured/theseus_accelerator_parity_manifest_import_prototype_backed.json` | The bounded sanitized Project Theseus accelerator parity manifest import records trigger state `GREEN`, 7 of 7 surfaces OK, 7 MLX report summaries, 4 Metal report summaries, 4 artifact manifests, 4 scheduler-canary surfaces, zero hard failures, zero explicit guardrail gaps, zero public training rows, zero external inference calls, no teacher use, locked model promotion and production routing, and nine expected-invalid controls. | No Project Theseus chapter core claim, clean live Project Theseus replay, full CUDA/MLX/Metal parity, production scheduler routing, model promotion, benchmark performance, model quality, safety, alignment, transfer, deployment readiness, or ASI claim moved. |
| `moral-uncertainty-and-value-conflict.theseus_governance_rights_receipt_suite_import` | `prototype-backed` | `docs/theseus_governance_rights_receipt_suite_import.md`; `evidence_transitions/v1_x_measured/theseus_governance_rights_receipt_suite_import_prototype_backed.json` | The bounded sanitized Project Theseus governance-rights receipt suite import records 4/4 governance-right fixtures, 4/4 constitutional-predicate fixtures, four governance-right records, four constitutional-predicate records, eight evidence-transition records, eight artifact-graph records, eight failure-boundary records, zero public training rows, zero external inference calls, and seven expected-invalid controls. | No Moral Uncertainty and Contestable Governance chapter core claim, Constitutional Alignment chapter core claim, legal rights, institutional governance, moral correctness, reviewer independence, export usability, safe fork execution, deployed runtime enforcement, clean live Project Theseus replay, safety, alignment, transfer, deployment readiness, or ASI claim moved. |
| `resource-economics.simulation_fidelity_receipt_suite_import` | `prototype-backed` | `docs/theseus_simulation_fidelity_receipt_suite_import.md`; `evidence_transitions/v1_x_measured/theseus_simulation_fidelity_receipt_suite_import_prototype_backed.json` | The bounded sanitized Project Theseus simulation-fidelity receipt suite import records 5/5 passed fixture scenarios, six simulation contract records, six fidelity records, six world-adapter receipts, six evidence-transition records, six failure-boundary records, one blocked transfer, one downgraded claim, one scenario-only record, zero public training rows, zero external inference calls, and seven expected-invalid controls. | No Resource Economics chapter core claim, simulator adequacy, physical feasibility, benchmark transfer, native KV parity, deployment readiness, live simulator behavior, clean live Project Theseus replay, model quality, economic outcome, learned generation, safety, alignment, transfer, deployment readiness, or ASI claim moved. |
| `resource-economics.theseus_rlds_minari_trace_export_import` | `prototype-backed` | `docs/theseus_rlds_minari_trace_export_import.md`; `evidence_transitions/v1_x_measured/theseus_rlds_minari_trace_export_import_prototype_backed.json` | The bounded sanitized Project Theseus RLDS/Minari trace-export import records one READY export manifest, three declared formats, seven declared fields, required license metadata, required replay smoke, zero public training rows, zero external inference calls, no copied episode payload, and seven expected-invalid controls. | No Resource Economics chapter core claim, RLDS dataset correctness, Minari dataset quality, simulator adequacy, replay success, physical feasibility, benchmark transfer, clean live Project Theseus replay, model quality, economic outcome, deployment readiness, safety, alignment, transfer, or ASI claim moved. |

## Accepted Terminal Full-Attempt Dispositions

These records reached an honest terminal campaign outcome but did not earn an
upward support transition. They are shown separately from narrow upward
transitions and from infrastructure side-lane gates.

| Claim ID | Terminal disposition | Evidence packet | What the attempt established | What remains blocked |
|---|---|---|---|---|
| `routing-deliberation.heldout-local-policy-effect` | `mixed_bounded_effect_no_promotion` | `docs/p4_routing_deliberation_campaign.md`; `evidence_transitions/post_v2_3/routing_deliberation_mixed_no_promotion.json` | On one authored 32-task local held-out workload, the full reflexive policy improved route correctness and useful outcomes relative to LLM-first, but produced two unsafe outputs versus zero and did not establish useful verifier-gated deliberation repair. | Safe or generally useful routing, useful test-time scaling, external evaluator independence, transfer, deployment, SOTA, AGI, ASI, and every chapter-core support transition remain blocked. |
| `update-unlearning.full-state-local-axis-separation` | `claim_narrowed_after_full_attempt` | `docs/p4_update_unlearning_campaign.md`; `evidence_transitions/post_v2_3/update_unlearning_claim_narrowed_after_full_attempt.json` | After preserving two failed instruments, one terminal five-seed/seven-arm local campaign separated eight nonfungible claim axes, observed 35/35 exact 24-surface local rollbacks, a small `0.01222040` movement toward deletion retraining, reduced one internal membership-attack advantage by `0.02511111`, and invalidated five local descendants. | Language-model weight unlearning, zero causal influence, privacy, legal compliance, total storage or backup erasure, external-descendant closure, transfer, production rollback, SOTA, AGI, ASI, and every chapter-core support transition remain blocked. |

## Accepted No-Promotion Side-Lane Decisions

| Claim ID | Support-state effect | Evidence packet | What the decision records | What stays blocked |
|---|---|---|---|---|
| `residual-verifier-capacity.heldout-pressure-effect` | `blocks_promotion` | `docs/p4_residual_verifier_capacity_campaign.md`; `evidence_transitions/post_v2_3/residual_verifier_terminal_instrument_failure.json` | Three prospectively versioned repairs used 18 sacrificial tasks and 36 exact local Qwen3-8B calls. The terminal path reached 6/6 eligibility accuracy, 3/3 clean releases, perfect required-residual recall, and zero false reassurance, but only 3/6 extraction objects met the frozen action/check schema; all fifteen held-out tasks remained sealed and no claim attempt occurred. | Residual honesty under pressure, verifier competence, useful-throughput improvement, natural-task effect, safety, transfer, deployment, SOTA, AGI, ASI, publication, release, and every chapter-core support transition remain blocked. |
| `artifact-graphs.epistemic_tcb_fixture` | `blocks_promotion` | `docs/epistemic_trusted_computing_base_fixture.md`; `evidence_transitions/v1_x_measured/artifact_epistemic_tcb_fixture_no_change.json` | A bounded epistemic trusted-computing-base fixture accepts three finite trust-base records and rejects controls for missing roots of trust, same-component verifier laundering, unbounded trust propagation, missing recursion stops, erased outside-TCB residuals, and support promotion from trust-base shape. | Verifier correctness, deployed trust-base behavior, audit-log durability, policy correctness, open-world receipt faithfulness, source interpretation, benchmark performance, model quality, safety, ASI, and Artifact Graphs chapter-core promotion stay blocked. |
| `artifact-graphs.record_reality_sequence_bridge` | `blocks_promotion` | `docs/artifact_graph_record_reality_sequence.md`; `evidence_transitions/v1_x_measured/artifact_record_reality_sequence_no_change.json` | A bounded record-reality sequence accepts one stale/partial/fresh replay path and rejects controls for stale-certificate support movement, restoration without fresh replay/provenance, missing non-claims, and support review without replay-validated transaction state. | Deployed artifact-graph behavior, deployed replay, audit durability, open-world receipt faithfulness, verifier correctness, provenance completeness, source interpretation, benchmark performance, model quality, safety, ASI, and Artifact Graphs chapter-core promotion stay blocked. |
| `artifact-graphs.receipt_faithfulness_adversarial_fixture` | `blocks_promotion` | `docs/receipt_faithfulness_adversarial_fixture.md`; `evidence_transitions/v1_x_measured/artifact_receipt_faithfulness_no_change.json` | The bounded receipt-faithfulness fixture accepts three record-only or blocked receipt cases and rejects six controls for shape-valid reality-false receipts, ignored trap failures, missing independent cross-checks, same-component self-check laundering, unbounded attestation, and support promotion from receipt shape. | Open-world receipt faithfulness, verifier independence, attestation-service correctness, deployed audit behavior, provenance completeness, source interpretation, benchmark performance, model quality, safety, ASI, and Artifact Graphs chapter-core promotion stay blocked. |
| `artifact-graphs.receipt_repository_audit_challenge` | `blocks_promotion` | `docs/receipt_repository_audit.md`; `evidence_transitions/v1_x_measured/artifact_receipt_repository_audit_challenge_no_change.json` | The receipt repository audit accepts four selected repository receipt surfaces, checks 55 tracked artifact digests plus external Circle fingerprints, and the deterministic challenge accepts three tracked-digest responses plus one Circle fingerprint while rejecting ten audit/challenge mutation controls. | Deployed artifact-graph behavior, deployed attestation, deployed audit durability, open-world receipt faithfulness, verifier correctness, provenance completeness, external project truth, benchmark performance, model quality, safety, ASI, and Artifact Graphs chapter-core promotion stay blocked. |
| `artifact-graphs.live_artifact_attestation_probe` | `blocks_promotion` | `docs/artifact_live_attestation_probe.md`; `evidence_transitions/v1_x_measured/artifact_live_attestation_no_change.json` | One current produced artifact is checked through filesystem bytes, git object bytes, command replay, independent observer routes, a trap receipt, attestation limits, and seven mutation controls. | Deployed attestation, deployed audit behavior, open-world receipt faithfulness, verifier correctness, external project truth, and Artifact Graphs chapter-core promotion stay blocked until deployed or externally reviewable attestation/audit traces and independent review exist. |
| `artifact-graphs.randomized_artifact_attestation_audit` | `blocks_promotion` | `docs/artifact_randomized_attestation_audit.md`; `evidence_transitions/v1_x_measured/artifact_randomized_attestation_no_change.json` | A deterministic pseudo-random sample of four public-safe repository result artifacts is checked through filesystem bytes, git object bytes, command replay, wrong-digest trap receipts, attestation limits, and mutation controls. | Deployed or externally reviewable attestation, open-world receipt faithfulness, verifier correctness, provenance completeness, external artifact truth, and Artifact Graphs chapter-core promotion stay blocked. |
| `artifact-graphs.github_pages_ci_attestation` | `blocks_promotion` | `docs/artifact_github_pages_ci_attestation.md`; `evidence_transitions/v1_x_measured/artifact_github_pages_ci_attestation_no_change.json` | A GitHub Actions `Publish Quarto site` run for one `main` commit is captured as an external CI-service record with successful source validation, Lean build, HTML render, live Human view validation, browser smoke testing, Pages artifact upload, and deploy. | Independent external human review, deployed attestation, open-world receipt faithfulness, verifier correctness, reader release approval, future-commit status, and Artifact Graphs chapter-core promotion stay blocked. |
| `artifact-graphs.public_site_record_reality_attestation` | `blocks_promotion` | `docs/artifact_public_site_attestation.md`; `evidence_transitions/v1_x_measured/artifact_public_site_attestation_no_change.json` | The public GitHub Pages Artifact Graphs chapter is fetched over HTTPS and checked for served HTML digest, record-reality ladder, epistemic-TCB boundary, CI-attestation section, live/randomized attestation references, and non-claim boundary fragments. | Independent external human review, deployed attestation behavior, open-world receipt faithfulness, verifier correctness, reader release approval, future-commit status, and Artifact Graphs chapter-core promotion stay blocked. |
| `circle-calculus.public_consumer_gate` | `blocks_promotion` | `docs/circle_public_replay_consumer_gate.md`; `evidence_transitions/v1_x_measured/circle_public_consumer_gate_no_change.json` | The ASI-side public consumer gate validates one pinned `CC-AI-CONTRACT-ROPE-001` receipt fixture by digest, theorem IDs, deterministic fields, and four expected-invalid overclaim controls. | Proof-contract transport, model quality, context length, speed, memory scaling, transfer, safety, ASI, and Circle chapter-core promotion stay blocked until stronger replay, transport, workload, baseline, metric, and review artifacts exist. |
| `circle-calculus.contract_pack_archive` | `blocks_promotion` | `docs/circle_contract_pack_archive.md`; `evidence_transitions/v1_x_measured/circle_contract_pack_archive_no_change.json` | The Circle contract-pack archive validates a public-safe generated pack snapshot from commit `63b0f511` with 9 archived contracts, 4 acceptance-policy receipts, `public_safe_fixture` status, pack fingerprint `df673f8a661fc89a26372685986c92f2221aaa617d6738fce5c2a76bd5d0eeae`, raw pack digest `b5488c93109ef120b97fdea7bd5d5605f32b2618c6cbfb9dde9a3328652551c4`, raw report digest `f1671f5cecdee311185f7e4508b21c139d4ae4bd1fa9610a12827f4d31c7985a`, and five expected-invalid controls. | Does not promote any chapter core claim, does not create a support-state transition, does not rerun Circle Lean, does not prove deployed proof-contract transport, and does not prove model quality, context length, runtime speed, memory scaling, deployment safety, transfer, safety, or ASI; Circle chapter-core promotion stays blocked. |
| `circle-calculus.kv_cache_ring_buffer_receipt_slice` | `blocks_promotion` | `docs/circle_kv_cache_receipt_slice.md`; `evidence_transitions/v1_x_measured/circle_kv_cache_receipt_no_change.json` | The local external Circle KV-cache receipt slice records commit `63b0f511`, `CC-AI-CONTRACT-KV-001`, kind `kv_cache_ring_buffer`, theorem IDs `AIM-T0103`, `AIM-T0104`, `AIM-T0149`, `theorem_count=54`, recommendations `KV-DROP-STALE-REQUEST-TOKEN` and `KV-USE-SINK-ROLLING-WINDOW-REQUEST`, fingerprint `bfebf150ce45d1eb124ea553bf2ba8c62008751ebec9f8600b83cc09e0526a46`, `stale_probe_first_stale_token=12`, `sink_tokens_retained_by_policy=true`, `sink_window_exact_policy=true`, `sink_window_tokens_distinct=true`, `sink_prefix_disjoint_from_live_window=true`, `sink_tokens_outside_ordinary_rolling_window=true`, and `5 passed in 1.27s`. | Does not promote any chapter core claim, does not create a support-state transition, and does not prove deployed KV-cache behavior, serving throughput, memory savings, paging correctness, retrieval quality, model quality, context length, deployment safety, transfer, or ASI; Coil/Circle chapter-core promotion stays blocked. |
| `circle-calculus.recurrence_schedule_receipt_slice` | `blocks_promotion` | `docs/circle_recurrence_receipt_slice.md`; `evidence_transitions/v1_x_measured/circle_recurrence_receipt_no_change.json` | The local external Circle recurrence receipt slice records commit `63b0f511`, `CC-AI-CONTRACT-RECURRENCE-001`, kind `recurrence_schedule`, theorem IDs `AIM-T0026`, `AIM-T0130`, `AIM-T0159`, `theorem_count=64`, recommendations `RECURRENCE-USE-ACTIVE-TOKEN-WORK-SCHEDULE` and `RECURRENCE-REUSE-WHOLE-PERIOD-SHIFT`, fingerprint `571edd5dce4f7b64441806de323295218a3e2293b3b540dd4772ba34b9371515`, `active_token_count_trace=[8, 6, 4, 2, 1]`, `inactive_token_count_trace=[0, 2, 4, 6, 7]`, `total_active_token_work=21`, `total_inactive_token_work=19`, `scheduled_work_saving=19`, `post_period_multi_extension_scheduled_work_saving=43`, `periodic_shift_required_steps_invariant=true`, `periodic_shift_active_at_step_invariant=true`, and `2 passed in 2.37s`. | Does not promote any chapter core claim, does not create a support-state transition, and does not prove deployed recurrence behavior, reasoning quality, retrieval quality, learned-memory behavior, convergence, model quality, context length, deployment safety, transfer, or ASI; Coil/Circle chapter-core promotion stays blocked. |
| `circle-calculus.sparse_attention_receipt_slice` | `blocks_promotion` | `docs/circle_sparse_attention_receipt_slice.md`; `evidence_transitions/v1_x_measured/circle_sparse_attention_receipt_no_change.json` | The local external Circle sparse-attention receipt slice records commit `63b0f511`, `CC-AI-CONTRACT-SPARSE-001`, kind `sparse_attention_coverage`, theorem IDs `AIT-T0104`, `AIT-T0172`, `theorem_count=141`, recommendations `SPARSE-LOCAL-FIRST-INTERVAL-REPAIR` and `SPARSE-DENSE-LOCAL-COMPLETE-FALLBACK`, fingerprint `c23809cef9b821b1e4f9cabf53fcac724a0757bf3f86594e1d12710fe0cd9ec1`, `coverage_complete=false`, `first_uncovered_lag=5`, `uncovered_lag_count=109`, `covered_lag_count=10`, `complete_repair_window=119`, `complete_repair_window_additional_local_slots=115`, `complete_repair_window_minimal_for_declared_stride_family=true`, `complete_repair_window_minimal_witness_lag=119`, `interval_repair_plan_step_count=6`, `lag_collision_pair_count=0`, `query_collision_pair_count=0`, and `10 passed in 1.87s`. | Does not promote any chapter core claim, does not create a support-state transition, does not prove sparse-attention coverage success, and does not prove deployed sparse-attention behavior, retrieval quality, long-context quality, model quality, speed, memory savings, deployment safety, transfer, or ASI; Coil/Circle chapter-core promotion stays blocked. |
| `circle-calculus.strided_fanout_receipt_slice` | `blocks_promotion` | `docs/circle_strided_fanout_receipt_slice.md`; `evidence_transitions/v1_x_measured/circle_strided_fanout_receipt_no_change.json` | The local external Circle strided fanout receipt slice records commit `63b0f511`, `CC-AI-CONTRACT-FANOUT-001`, kind `strided_candidate_fanout`, theorem IDs `AIT-T0001`, `AIT-T0002`, `AIT-T0003`, `AIT-T0173`, `theorem_count=4`, recommendations `FANOUT-USE-FULL-COVERAGE-STRIDE-CYCLE` and `FANOUT-AUDIT-DUPLICATE-COLLAPSED-BUDGET`, fingerprint `d4c878563747da9c9f1f55cd689f04e2a0a8e31ce9429a138341ec4e27ee3799`, `context_length=12`, `stride=5`, `gcd=1`, `predicted_reach=12`, `full_coverage=true`, `candidate_budget=12`, `unique_candidate_count=12`, `effective_candidate_budget=12`, `duplicate_count=0`, `candidate_budget_shortfall=0`, `effective_budget_matches_unique_candidates=true`, `effective_budget_reaches_predicted_reach=true`, `3 passed in 4.65s`, and `1 passed in 2.77s`. | Does not promote any chapter core claim, does not create a support-state transition, and does not prove search quality, retrieval quality, routing quality, sparse-attention quality, model quality, throughput, latency, runtime speed, memory savings, deployment safety, transfer, optimal scheduling, or ASI; Coil/Circle chapter-core promotion stays blocked. |
| `circle-calculus.cyclic_mixer_receipt_slice` | `blocks_promotion` | `docs/circle_cyclic_mixer_receipt_slice.md`; `evidence_transitions/v1_x_measured/circle_cyclic_mixer_receipt_no_change.json` | The local external Circle cyclic-mixer receipt slice records commit `63b0f511`, `CC-AI-CONTRACT-MIXER-001`, kind `circulant_block_cyclic_mixer`, theorem IDs `AIT-T0006`, `AIT-T0007`, `AIT-T0008`, `AIT-T0009`, `AIRA-T0001`, `AIRA-T0002`, `AIRA-T0004`, `theorem_count=7`, recommendations `MIXER-AUDIT-CIRCULANT-DENSE-PARITY` and `MIXER-AUDIT-BLOCK-CYCLIC-PARAMETER-BUDGET`, fingerprint `b3e3e0cf420d9e8e79a28a55ef8322f9a214c8d5a957dd8b06e5e5373c684ea5`, `max_abs_dense_delta=0`, `dense_parameters=64`, `circulant_parameters=8`, `circulant_parameter_ratio=0.125`, `dense_adapter_parameters=2048`, `lora_parameters=576`, `block_cyclic_parameters=128`, `block_to_dense_ratio=0.0625`, `3 passed in 2.49s`, and `1 passed in 1.47s`. | Does not promote any chapter core claim, does not create a support-state transition, does not prove cyclic-mixer model quality, does not prove runtime speed, does not prove memory scaling, hardware efficiency, training stability, deployment readiness, transfer, benchmark performance, or ASI; CoilRA/Circle chapter-core promotion stays blocked. |
| `circle-calculus.multicoil_phase_receipt_slice` | `blocks_promotion` | `docs/circle_multicoil_phase_receipt_slice.md`; `evidence_transitions/v1_x_measured/circle_multicoil_phase_receipt_no_change.json` | The local external Circle MultiCoil phase receipt slice records commit `63b0f511`, `CC-AI-CONTRACT-PHASE-FEATURE-001`, kind `multicoil_phase_feature`, theorem IDs `AIA-T0001`, `AIA-T0002`, `AIA-T0004`, `AIT-T0004`, `AIT-T0005`, `theorem_count=5`, recommendations `PHASE-USE-JOINT-REPEAT-HORIZON` and `PHASE-AUDIT-RELATIVE-SHIFT-INVARIANT`, fingerprint `4b562beab64ec863903e4267f50c90049f0d3fa612f6c1bb2f06ad07e821ffd7`, `periods=[5, 7]`, `phase_tuple=[2, 2]`, `shifted_phase_tuple=[2, 2]`, `joint_repeat_horizon=35`, `relative_phase=3`, `shifted_relative_phase=3`, `relative_phase_invariant=true`, `3 passed in 2.99s`, and `1 passed in 1.76s`. | Does not promote any chapter core claim, does not create a support-state transition, does not prove MultiCoil, RoPE, attention, retrieval, or model quality, and does not prove context length, runtime speed, memory scaling, hardware efficiency, training stability, deployment readiness, transfer, benchmark performance, or ASI; CoilRA/Circle chapter-core promotion stays blocked. |
| `planning.scheduler_state_probe` | `blocks_promotion` | `docs/planning_scheduler_state_probe.md`; `evidence_transitions/v1_x_measured/planning_scheduler_state_probe_no_change.json` | The deterministic Planning scheduler-state probe validates two synthetic scheduler traces and rejects seven controls for blocked-node dispatch, ready-without-context dispatch, failed-adequacy route selection, conflicting merge acceptance, replanning authority erasure, dependency-cycle acceptance, and hidden repair, human-review, failed-attempt, or residual cost erasure. | Deployed planner behavior, deployed scheduler behavior, decomposition quality, context-demand prediction, route quality, selected-tier adequacy, scheduler optimality, runtime replanning, live feedback handling, benchmark performance, model quality, safety, ASI, and Planning chapter-core promotion stay blocked. |
| `planning.runtime_replan_delta_audit` | `blocks_promotion` | `docs/planning_runtime_replan_delta_audit.md`; `evidence_transitions/v1_x_measured/planning_runtime_replan_delta_no_change.json` | The deterministic Planning runtime-replan delta audit validates two synthetic runtime-replan traces and rejects nine controls for authority widening, stop-condition erasure, unaffected-node reruns without dependency impact, missing residual ownership, missing context or verification deltas, blocked-authority dispatch, support-promotion overclaim, and missing non-claim boundaries. | Deployed planner behavior, runtime scheduler behavior, decomposition quality, context-demand prediction, route quality, selected-tier adequacy, scheduler optimality, live feedback handling, deployed runtime replanning, benchmark performance, model quality, safety, ASI, and Planning chapter-core promotion stay blocked. |
| `fast-generation-architectures.public_safe_task_bundle_accounting` | `blocks_promotion` | `docs/fast_generation_task_bundle.md`; `evidence_transitions/v1_x_measured/fast_generation_task_bundle_no_change.json` | A deterministic public-safe task bundle keeps accepted-output accounting, baseline comparison, verifier/fallback/residual boundaries, and latency-only rejection visible without treating the task bundle as model-speed evidence. | Model generation speed, useful-solution-per-second performance, serving throughput, route-selector adequacy, benchmark performance, model quality, deployed routing, and Fast Generation chapter-core promotion stay blocked. |
| `runtime-adapters.local_effect_replay_probe` | `blocks_promotion` | `docs/runtime_adapter_effect_probe.md`; `evidence_transitions/v1_x_measured/runtime_adapter_effect_probe_no_change.json` | The Runtime adapter effect replay probe runs one generated public-safe temp-file write outside the repository, records pre/post/rollback hashes, verifies rollback-exact restoration, and rejects missing-permission plus expired-approval controls before mutation. | Deployed adapter behavior, sandbox isolation, approval-service behavior, secret-handle safety, revocation propagation, policy-enforcement correctness, rollback-service behavior, benchmark performance, runtime security, model behavior, and Runtime Adapters chapter-core promotion stay blocked. |
| `runtime-adapters.adversarial_boundary_probe` | `blocks_promotion` | `docs/runtime_adapter_adversarial_boundary_probe.md`; `evidence_transitions/v1_x_measured/runtime_adapter_adversarial_boundary_no_change.json` | The Runtime adapter adversarial boundary probe records two valid synthetic boundary reviews and twelve expected-invalid controls for parentage, authority ceilings, approval scope/expiry, sandbox escape, secret materialization, rollback, receipt, audit, support-state, and non-claim boundaries. | Deployed adapter behavior, sandbox isolation, approval-service behavior, secret-handle safety, policy-enforcement correctness, rollback-service behavior, revocation propagation, security review, benchmark performance, runtime security, and Runtime Adapters chapter-core promotion stay blocked. |
| `runtime-adapters.human_oversight_degradation` | `blocks_promotion` | `docs/human_oversight_degradation_fixture.md`; `evidence_transitions/v1_x_measured/human_oversight_degradation_no_change.json` | The finite human-oversight degradation fixture validates three bounded synthetic approval-review records and rejects seven controls for missing reviewer qualification, fatigued approval, rubber-stamping, automation-bias contradiction, alarm fatigue, support-promotion overclaim, and missing non-claim boundaries. | Approval-service quality, reviewer correctness, deployed human-factors behavior, runtime-adapter safety, alert quality, approval-workflow evidence, and Runtime Adapters chapter-core promotion stay blocked until public-safe deployed or externally reviewable approval-workflow traces, reviewer-load fields, independent-evidence checks, reviewer rotation or delay behavior, alert-quality evidence, privacy-boundary handling, residuals, and independent review exist. |
| `personal-compute-hives.partitioned_authority_fixture` | `blocks_promotion` | `docs/partitioned_authority_fixture.md`; `evidence_transitions/v1_x_measured/partitioned_authority_fixture_no_change.json` | The finite partitioned-authority fixture validates three hives/runtime boundary records and rejects six controls for stale-grant dispatch, mutation after unseen revocation, grant/effect race residual ownership, missing no-mutation evidence, support-state promotion, and missing non-claim boundaries. | Deployed partition tolerance, distributed consensus, availability, runtime-adapter enforcement, revocation propagation, hive scheduler behavior, rented-node sandbox behavior, family-governance behavior, network-overlay behavior, security, privacy, benchmark performance, and Personal Compute Hives or Runtime Adapters chapter-core promotion stay blocked. |
| `system-boundaries.authority_revocation_trace` | `blocks_promotion` | `docs/authority_revocation_trace.md`; `evidence_transitions/v1_x_measured/authority_revocation_trace_no_change.json` | The cross-artifact authority revocation trace checks over-ceiling denial, revoked authority receipt blocking, expired approval no-mutation evidence, SCIF inactive approval blocking, and reference-trace authority blockers across existing repository artifacts. | Deployed authorization enforcement, live revocation propagation, runtime adapter enforcement, approval-service quality, sandbox isolation, tool-wrapper security, benchmark performance, safety, and System Boundaries chapter-core promotion stay blocked. |
| `rankfold-neuralfold.local_artifact_import_metadata` | `blocks_promotion` | `docs/rankfold_artifact_import.md`; `evidence_transitions/v1_x_measured/rankfold_artifact_import_no_change.json` | Three existing local `.rfa` archive observations are imported as public-safe metadata with decoded digest consensus, verifier summaries, and `NEURAL0` inspect metadata boundaries. | RankFold codec correctness, deterministic decoder correctness beyond the recorded local observations, compression from source input, benchmark performance, downstream utility, fallback execution, deployed compression behavior, and RankFold chapter-core promotion stay blocked. |
| `rankfold-neuralfold.public_safe_replay_probe` | `blocks_promotion` | `docs/rankfold_public_safe_probe.md`; `evidence_transitions/v1_x_measured/rankfold_public_safe_replay_probe_no_change.json` | A fresh local RankFold pack/verify/list/unpack replay over a synthetic public-safe fixture records RAW0 roundtrip exactness, no compression advantage, a license-disabled NeuralFold boundary, and a corrupt-archive negative control. | NeuralFold compression, compression advantage, codec correctness, downstream utility, fallback execution, deployed compression behavior, model quality, benchmark performance, and RankFold chapter-core promotion stay blocked. |
| `compact-generative-systems.seed_rule_exact_regeneration_receipt_slice` | `blocks_promotion` | `docs/circle_seed_rule_receipt_slice.md`; `evidence_transitions/v1_x_measured/circle_seed_rule_receipt_no_change.json` | The local external Circle seed-rule receipt slice records commit `63b0f511`, `CC-AI-CONTRACT-SEED-RULE-001`, kind `seed_rule_exact_regeneration`, theorem IDs `GEN-T0001`, `GEN-T0040`, `GEN-T0041`, `GEN-T0046`, `GEN-T0048`, `GEN-T0050`, `theorem_count=32`, recommendations `SEED-RULE-USE-EXACT-REGENERATION-RECIPE` and `SEED-RULE-SELECT-BOUNDED-SHORTER-CANDIDATE`, fingerprint `836594a5f1d448900797e595cb98f0e476c0b9cbd7365fe333cf7ae2622f13c5`, `fixture_n=128`, `exact_regeneration=true`, `generator_length=383`, `explicit_length=454`, `storage_saving=71`, `bounded_search_candidate_count=3`, `bounded_search_exact_candidate_count=2`, `bounded_search_best_shorter_generator_shorter=true`, `2 passed in 4.52s`, and `1 passed in 2.80s`. | Does not promote any chapter core claim, does not create a support-state transition, and does not prove useful compression, codec correctness, semantic utility, deployed generator behavior, fallback execution, downstream utility, optimal search, model quality, context length, runtime speed, memory scaling, benchmark performance, transfer, safety, or ASI; Compact Generative Systems chapter-core promotion stays blocked. |
| `compact-generative-systems.residual_ledger_storage_replay` | `blocks_promotion` | `docs/residual_ledger_storage_replay.md`; `evidence_transitions/v1_x_measured/residual_ledger_storage_replay_no_change.json` | The bounded residual-ledger storage/replay fixture validates four append-only replay entries and rejects five controls for handoff-owner mismatch, missing discharge receipt or review, sequence gap, missing workload context, and support promotion from replay shape. | Deployed residual-ledger storage, live residual detection, safety, model quality, benchmark performance, residual observability, and Compact Generative Systems chapter-core promotion stay blocked until live or externally reviewable storage/replay evidence, public-safe workload context, adversarial replay behavior, and independent review exist. |
| `resource-economics.publication_pipeline_cost_profile` | `blocks_promotion` | `docs/resource_ci_cost_profile.md`; `evidence_transitions/v1_x_measured/resource_ci_cost_profile_no_change.json` | GitHub Pages publication-pipeline runs, failure classification, recovery boundary, and duration metrics are recorded as repository publication metadata only. | Production scheduler behavior, deployed resource governance, economic optimization, model quality, runtime enforcement, physical feasibility, and Resource Economics chapter-core promotion stay blocked. |
| `resource-economics.local_replay_probe` | `blocks_promotion` | `docs/resource_live_probe.md`; `evidence_transitions/v1_x_measured/resource_live_probe_no_change.json` | Five local Resource Economics validators are replayed with command-output and tracked-artifact digests to make the lane reproducible and drift-checkable. | Live workload quality, deployed scheduler behavior, runtime budget enforcement, production workload evidence, and Resource Economics chapter-core promotion stay blocked. |
| `resource-economics.synthetic_load_stability_route_selection` | `blocks_promotion` | `docs/resource_load_stability_probe.md`; `evidence_transitions/v1_x_measured/resource_load_stability_probe_no_change.json` | The broader load-stability probe records a finite synthetic workload, selected route, baseline, review-erasure negative control, residualized deferrals, and Lean/Python fixture alignment. | Real load stability, deployed scheduler behavior, TokenMana/PlanForge behavior, reviewer optimization, economic claims, and Resource Economics chapter-core promotion stay blocked. |
| `resource-economics.workflow_trace_dispatch_accounting` | `blocks_promotion` | `docs/resource_workflow_trace.md`; `evidence_transitions/v1_x_measured/resource_workflow_trace_no_change.json` | The workflow trace checks selected-route cost recomputation, high-risk-first ordering, protected review, verification minutes, displaced-cost residual ownership, over-budget rejection, and latency-only rejection. | Deployed scheduler behavior, route-search completeness, model quality, economic outcomes, physical-feasibility claims, and Resource Economics chapter-core promotion stay blocked. |
| `resource-economics.local_workload_quality_route_selection` | `blocks_promotion` | `docs/resource_workload_quality_probe.md`; `evidence_transitions/v1_x_measured/resource_workload_quality_probe_no_change.json` | The local workload-quality probe records a measured route-selection artifact with five samples, a passing quality predicate, and a cheaper no-op rejection. | Stable speedup, production workload quality, deployed scheduler behavior, model quality, economic outcomes, and Resource Economics chapter-core promotion stay blocked. |
| `routing-heads.synthetic_routing_decision_lease` | `blocks_promotion` | `docs/routing_decision_lease_harness.md`; `evidence_transitions/v1_x_measured/routing_decision_lease_no_change.json` | Synthetic route receipts reject overprivileged selection, authority widening, stale leases, missing readiness, missing residual ownership, and source-only MoECOT promotion at record level. | Learned-router quality, route-quality dominance, deployed authority enforcement, specialist quality, MoECOT replay, orchestration benchmark performance, runtime behavior, model quality, safety, and Routing chapter-core promotion stay blocked. |
| `readiness-gates.readiness_lifecycle_probe` | `blocks_promotion` | `docs/readiness_lifecycle_probe.md`; `evidence_transitions/v1_x_measured/readiness_lifecycle_probe_no_change.json` | The deterministic readiness lifecycle probe validates six finite lifecycle transitions and rejects twelve controls for non-forward jumps, missing fresh evidence, missing residual escrow, unsafe default readiness, quarantine leakage, missing terminal records, retired-state reuse, missing non-claims, and support promotion. | Deployed readiness-engine behavior, lifecycle execution, residual-ledger storage, live quarantine routing, gate-quality checks, terminal-state governance, rollback execution, runtime monitoring, MoECOT replay, benchmark performance, current module readiness, and Readiness Gates chapter-core promotion stay blocked. |
| `labor-os.typed_job_durable_lifecycle_probe` | `blocks_promotion` | `docs/typed_job_durable_lifecycle_probe.md`; `evidence_transitions/v1_x_measured/typed_job_durable_lifecycle_probe_no_change.json` | The deterministic typed-job durable lifecycle probe validates two synthetic durable lifecycle traces and rejects nine controls for retry idempotency, authority preservation, permission scope, expired-lease dispatch blocking, completion receipts, replay refs, residual ownership, non-claim boundaries, and support promotion. | Deployed scheduler behavior, durable workflow recovery, permission enforcement, approval-service behavior, adapter execution, completion-receipt service behavior, replay correctness, workflow traces, benchmark performance, model quality, safety, ASI, and Labor OS chapter-core promotion stay blocked. |
| `security-kernel.scif_sanitized_commit_replay` | `blocks_promotion` | `docs/security_scif_commit_probe.md`; `evidence_transitions/v1_x_measured/security_scif_commit_probe_no_change.json` | The SCIF sanitized commit replay probe records two valid local synthetic routes and six expected-invalid controls for secret canary leakage, live handle leakage, missing zeroization, overbroad context admission, unapproved destination or inactive approval, and missing residual ownership. | Deployed-kernel behavior, sandbox isolation, side-channel safety, prompt-injection containment, secret-handle safety, approval-service behavior, least-privilege context behavior, privacy, security, and Security Kernel chapter-core promotion stay blocked. |
| `project-theseus-as-report-first-implementation-reference.public_task_bundle_import_summary` | `blocks_promotion` | `docs/theseus_public_task_bundle_import.md`; `evidence_transitions/v1_x_measured/theseus_public_task_bundle_import_no_change.json` | The public task-bundle import preserves metadata-only task counts, gate counts, residuals, artifact gaps, dirty-checkout boundaries, and expected-invalid controls while rejecting clean-live-replay and model-quality overclaims. | Clean live Project Theseus replay, deployed Theseus behavior, model quality, benchmark superiority, generation speed, useful-solution-per-second improvement, self-evolution safety, and Project Theseus chapter-core promotion stay blocked. |
| `project-theseus-as-report-first-implementation-reference.book_to_theseus_crosswalk_pointer` | `blocks_promotion` | `docs/theseus_book_crosswalk_import.md`; `evidence_transitions/v1_x_measured/theseus_book_crosswalk_import_no_change.json` | The sanitized book-to-Theseus crosswalk import records 53 public-safe pointer rows, 20 backlog cards, 134 source-sync review decisions, 9 changed AI-book source files, zero private/raw payload exports, zero public training rows, zero external inference calls, and ten expected-invalid controls. | Clean live Project Theseus replay, artifact truth for referenced rows, deployment, model quality, benchmark performance, self-evolution safety, and Project Theseus chapter-core promotion stay blocked. |
| `project-theseus-as-report-first-implementation-reference.work_board_currentness_import` | `blocks_promotion` | `docs/theseus_work_board_import.md`; `evidence_transitions/v1_x_measured/theseus_work_board_import_no_change.json` | The sanitized work-board metadata import records 130 durable task rows, 412 event rows, 133 evidence rows, five SQLite tables, one execution-ledger row, four unattended-improvement rows, 72 feedback rows, a stale-snapshot boundary, zero public training rows, zero external inference calls, and ten expected-invalid controls. | Clean live Project Theseus replay, current board state, current dashboard state, deployment, model quality, benchmark performance, unattended safety, self-evolution safety, support-state movement, and Project Theseus chapter-core promotion stay blocked. |

## Live Claim-Surface Narrowing Records

The P5 mandatory terminal batch also accepts the no-promotion record
`capability-replacement-and-rollback.invariant.011` at
`evidence_transitions/post_v2_3/p5_rollback_outcomes_narrowed_after_full_attempt.json`.
It preserves separate artifact, digest, behavior, privacy-proxy, storage,
backup, and descendant outcomes while leaving service restart and external
compensation unexecuted. Its support state remains `argument` and its effect is
`blocks_promotion`.

| Claim surface | Record | What narrowed | Support-state effect |
|---|---|---|---|
| `non-core-evidence-ledger.chapter-core-count-surface` | `claim_revisions/v1_x/manifest_core_claim_count_narrowing.json` | The obsolete public count phrase "All 54 remain at `argument`" was narrowed to the then-current 44-chapter manifest state; the active manifest now has 87 core claims, all at `argument`, and current counts are regenerated from `book_structure.json` and `docs/core_claim_transition_coverage.md`. | None; this is a count-surface correction, not a demotion, refutation, or promotion. |

## How To Promote Anything Later

A future upward transition needs a separate accepted evidence-transition record
that names the exact claim, artifact, command or replay path, limitations,
counterevidence, downgrade triggers, non-claims, and support-state effect.
External citations, source notes, green validators, local project summaries, or
reader-quality edits do not promote a claim by themselves.

## Validation

The ledger is checked by `scripts/validate_non_core_evidence_ledger.py`. That
validator reads the accepted transition records under
`evidence_transitions/v1_0_measured/` and `evidence_transitions/v1_x_measured/`,
checks that all 19 current non-core upward claims are listed here, derives and
checks all 57 accepted `blocks_promotion` side-lane decisions, checks the live
claim-surface narrowing record, checks the chapter-core non-promotion boundary,
and checks that the public entry surfaces link to this ledger. The live
revision record is also checked by `scripts/validate_claim_revision_records.py`.

## Accepted Post-v2 No-Change Programs

| Claim ID | Support-state effect | Evidence packet | What the decision records | What stays blocked |
|---|---|---|---|---|
| `post_v2_governed_work_flagship.bounded_matched_local_result` | `blocks_promotion` | `docs/post_v2_governed_work_flagship.md`; `evidence_transitions/post_v2/governed_work_flagship_no_change.json` | Sixteen retained model plan/code runs, matched Git worktrees, independent subprocess probes, receipt/path checks, residuals, ten rollbacks, eight exact and two failed, zero governed releases. | Production transfer, deployed service behavior, external independence, verifier correctness, production economics, and all three affected core promotions. |
| `post_v2_routing_deliberation.bounded_matched_local_result` | `blocks_promotion` | `docs/post_v2_routing_deliberation.md`; `evidence_transitions/post_v2/routing_deliberation_no_change.json` | Three-seed held-out routing and deliberation records, matched caps, 15 fixed-step harms, and explicit zero-use fallback/abstention. | Ambiguous-route validity, fallback calibration, trained-specialist interference, language-model deliberation, verifier correctness, transfer, and both core promotions. |
| `post_v2_update_causality.bounded_real_mutation_result` | `blocks_promotion` | `docs/post_v2_update_causality.md`; `evidence_transitions/post_v2/update_causality_no_change.json` | Real parameter/checkpoint/output changes, forgetting, deletion-cohort differences, best/final authority, three exact rollbacks, and nine invalidated descendant arms. | Production learning/unlearning, storage erasure, feedback optimization, open-ended improvement, recursive improvement, external independence, and four core promotions. |

## Accepted Post-v2.1 Bounded and No-Change Results

| Claim ID | Support-state effect | Evidence packet | What the decision records | What stays blocked |
|---|---|---|---|---|
| `post_v2_1.ambiguous_routing.bounded_result` | `blocks_promotion` | `evidence_transitions/post_v2_1/ambiguous_routing_narrow.json` | The learned router selected 59/60 routes and exercised fallback, abstention, and clarification, but all 360 substantive candidates were wrong; apparent utility came entirely from correct non-answer actions. | Substantive answer utility, evaluator validity, multi-model/natural transfer, production safety, latency, and cost. |
| `post_v2_1.full_state_rollback.bounded_result` | `blocks_promotion` | `evidence_transitions/post_v2_1/full_state_rollback_narrow.json` | All 15 seed-arm transactions restored all 24 declared model, optimizer, scheduler, RNG, cache, checkpoint, backup, and descendant surfaces exactly. | Heterogeneous runtimes, remote effects/backups, production recovery objectives, monitoring, and independent replay. |
| `post_v2_1.full_state_update.no_change_result` | `blocks_promotion` | `evidence_transitions/post_v2_1/full_state_update_no_change.json` | Prospective checkpoint authority and retained-task bounds held, but zero of nine eligible challenger arms reached the registered 0.05 target-utility gain. | Replicated gain, production-scale update/monitoring, feedback learning, open-endedness, recursion, and core promotion. |
| `post_v2_1.governed_usefulness_rollback.bounded_result` | `blocks_promotion` | `evidence_transitions/post_v2_1/governed_usefulness_rollback_narrow.json` | Governance reduced primary unsafe releases from 24/36 to 0/36, while useful release was 2/36 and attack-control exact rollback 32/36, below both frozen thresholds. | Useful governed throughput, open-system effect-complete rollback, independent verifier validity, production transfer, and acceptable governance cost. |
| `post_v2_1.real_model_deliberation.no_change_result` | `blocks_promotion` | `evidence_transitions/post_v2_1/real_model_deliberation_no_change.json` | All five real-model deliberation arms ended at 0/60 correct; adaptive stopping exhausted five candidates on every request, so corruption reduction was not estimable. | Matched-budget adaptive benefit, initial-correct corruption estimates, replication, independent evaluator validity, and transfer. |
| `post_v2_1.unlearning_causality.narrow_result` | `blocks_promotion` | `evidence_transitions/post_v2_1/unlearning_causality_narrow.json` | Deletion-aware retraining changed cohort behavior by 4/0/1 and propagated lineage invalidation; influence reduction remained a confidence proxy and storage erasure was false at every seed. | Causal influence removal, privacy/member assessment, descendant-wide propagation, verified backup/storage erasure, and core promotion. |

## Adjudicated QCSA v2.3 Non-Core Claims

The later KERC campaign retains one historical `refuted` transition label for
`kerc.broad_matched_total_system_efficiency`: all compared cores tied at
`0.500000` while complete Kernel packets averaged `714.0` bytes against a
`73.25`-byte best simple representation. The competence audit classifies the
run as N1 because chance-level task performance and implementation defects mean
the tested implementation was inadequate. The observation survives; broader
KERC, Kernel English, learned compilation, hierarchical residuals, and the
parent architecture remain untested. The immutable record is
`evidence_transitions/post_v2_3/kerc_broad_efficiency_refuted.json`, and the
current interpretation is bound in
`evidence_quality/negative_result_rehabilitation.json`.

These rows are the terminal P4 review of the ten evaluation recommendations
plus the implementation-only and vertical-reference candidates. The five
upward transitions apply only to their exact synthetic non-core claims. Two
rows retain historical `refuted` labels but are N2 proxy/regime evidence, not
exact competent refutations. Every other row remains at `argument`; every
chapter core remains at `argument`.

| Claim ID | Accepted state/effect | Transition | What the decision records | What stays blocked |
|---|---|---|---|---|
| `qcsa.plural_facets_exact_fixture_value` | `synthetic-test-backed` | `evidence_transitions/post_v2_3/qcsa_plural_facets_exact_fixture_value_promote.json` | Full QCSA resolved 1.000 of exact held-out objects while the no-plural-facets ablation resolved 0.916667 under the frozen 180-record corpus. | Natural-language, multilingual, multimodal, learned-representation, production, and chapter-core benefit. |
| `qcsa.identity_indirection_exact_migration_value` | `synthetic-test-backed` | `evidence_transitions/post_v2_3/qcsa_identity_indirection_exact_migration_value_promote.json` | Removing identity/address indirection reduced object accuracy to 0.900000 and migration compatibility to 0.400000 from 1.000/1.000. | Distributed registries, open-world identity, federation, security, and Virtual Context ABI core promotion. |
| `qcsa.certificate_authority_fields_exact_value` | `synthetic-test-backed` | `evidence_transitions/post_v2_3/qcsa_certificate_authority_fields_exact_value_promote.json` | Full QCSA had zero unsafe authority releases; the no-certificate/residual/authority ablation had nine across 180 records. | Runtime security, privacy, human-approval quality, general safety, and Runtime Adapters core promotion. |
| `qcsa.migration_compatibility_exact_value` | `synthetic-test-backed` | `evidence_transitions/post_v2_3/qcsa_migration_compatibility_exact_value_promote.json` | Removing migration compatibility reduced task-decision accuracy from 1.000 to 0.833333 and compatibility from 1.000 to 0.000. | Live descendants, distributed caches, production rollback, learning/unlearning, storage erasure, and Data Engines core promotion. |
| `qcsa.task_calibration_exact_result` | `synthetic-test-backed` | `evidence_transitions/post_v2_3/qcsa_task_calibration_exact_result_promote.json` | QCSA's exact task-decision Brier score was 0.082026 versus 0.298500 for the selected baseline, with selective risk 0 for both. | Object calibration, natural prevalence, non-ceiling tasks, learned models, external replication, and any chapter-core promotion. |
| `qcsa.exact_synthetic_matched_advantage` | historical `refuted`; N2 maximum inference | `evidence_transitions/post_v2_3/qcsa_exact_synthetic_matched_advantage_refute.json` | QCSA and the selected baseline both reached 1.000 task-decision accuracy while QCSA used 1.913386 times the operations, so both preregistered proxy gates failed. | Exact or broad refutation; natural corpora, learned models, independent evaluation, transfer, and other implementations remain untested. |
| `qcsa.active_questions_exact_fixture_value` | historical `refuted`; N2 maximum inference | `evidence_transitions/post_v2_3/qcsa_active_questions_exact_fixture_value_refute.json` | Removing active questions preserved 1.000 object and task-decision accuracy; the full system asked questions on 0.033333 of records without accuracy gain. | Exact or broad refutation; the saturated template workload, internal evaluator, and verifier-cost confound bound this to proxy/regime evidence. |
| `qcsa.semantic_round_trip_exact_preservation` | `blocks_promotion` | `evidence_transitions/post_v2_3/qcsa_semantic_round_trip_exact_preservation_narrow.json` | The internal independently implemented observer measured zero structural loss and zero disagreement for exact QCSA outputs. | Arbitrary semantic equivalence, independent replication, natural artifacts, and Cognitive Compilation core promotion. |
| `qcsa.governance_prevention_resource_tradeoff` | `blocks_promotion` | `evidence_transitions/post_v2_3/qcsa_governance_prevention_resource_tradeoff_narrow.json` | Exact prevention was 1.000000 versus 0.487179, but the 1.913386 operation ratio exceeded the preregistered 1.50 ceiling. | Acceptable production economics, incident-rate interpretation, governance efficacy, and Resource Economics core promotion. |
| `qcsa.open_world_or_production_transfer` | `blocks_promotion` | `evidence_transitions/post_v2_3/qcsa_open_world_or_production_transfer_no_change.json` | No natural workload, learned model, deployment, external replication, live effect, or prevalence study was run. | All open-world, production, safety, privacy, security, learned-model, AGI, and ASI claims. |
| `qcsa.reference_implementation_exact_contract_conformance` | `blocks_promotion` | `evidence_transitions/post_v2_3/qcsa_reference_implementation_exact_contract_conformance_narrow.json` | The deterministic reference package conforms to its exact schemas, digests, fixtures, and expected-invalid controls. | Treating implementation conformance as held-out evaluation, semantic quality, transfer, or Integrated Architecture core support. |
| `qcsa.vertical_reference_exact_reversible_trace` | `blocks_promotion` | `evidence_transitions/post_v2_3/qcsa_vertical_reference_exact_reversible_trace_narrow.json` | One 13-stage hand-authored trace completed a local temporary-file effect and exact declared rollback. | Open-system effect-complete rollback, learned routing, useful throughput, production transfer, and Integrated Architecture core promotion. |

## Preregistered Post-v2.3 Campaign Dispositions

Both campaigns completed their frozen call budgets with zero retries. Their
outputs are retained as protocol failures: all 36 calls exhausted the
256-token output cap; 34 ended inside unclosed reasoning blocks and two closed
reasoning without emitting a requested final JSON object. Raw reasoning text
is not treated as a final decision, and
zero unsafe releases with zero total releases is not treated as safety
evidence.

| Claim ID | Support-state effect | Evidence packet | What the decision records | What stays blocked |
|---|---|---|---|---|
| `post_v2_3.governance_tax_natural_work.bounded_result` | `blocks_promotion` | `docs/post_v2_3_campaign_results.md`; `evidence_transitions/post_v2_3/governance_tax_natural_work_no_change.json` | Twelve matched baseline/governed natural-work calls produced zero parseable final outputs and therefore zero releases in either route. Seven raw traces contained all criterion terms, but none was an admissible candidate. The separate local rollback harness restored all nine declared surfaces on 12/12 probes and detected 12/12 omission controls. | Useful-throughput and unsafe-release contrasts, refusal calibration, governance cost, broader model quality, production rollback, independent validity, transfer, and all affected chapter-core promotions. |
| `post_v2_3.residual_honesty_under_pressure.bounded_result` | `blocks_promotion` | `docs/post_v2_3_campaign_results.md`; `evidence_transitions/post_v2_3/residual_honesty_under_pressure_no_change.json` | Twenty-four frozen calls across six scenarios and four pressure conditions mentioned residual IDs only inside capped reasoning text; none emitted a parseable final decision, correct route, or reopen trigger. | Residual-disclosure quality, pressure robustness, correct routing, premature-closure effects, independent validity, production behavior, and both affected chapter-core promotions. |
| `resource_economics.governed_useful_throughput_under_natural_work` | `blocks_promotion` | `docs/post_v2_3_governance_tax_flagship_renewal_results.md`; `evidence_transitions/post_v2_3/governance_tax_natural_work_renewal_no_change.json` | The repaired protocol completed 32 candidate outputs and 64 calls. Only 2/32 candidates were independently correct; both routes had zero useful releases and zero unsafe releases; the governed route abstained 32/32 times; all 32 declared local rollbacks and omission controls passed. The immutable task file contains 9 families and 9 attacked tasks despite the frozen metadata's 8/8 declaration. | Useful-throughput advantage, unsafe-release reduction, production rollback, governance efficacy, model quality, external independence, safety, and every affected chapter-core promotion remain blocked. |

## Non-Claims

- This ledger does not create new evidence.
- This ledger does not demote, deprecate, or refute any chapter core claim.
- This ledger does not create an independent external review record.
- This ledger does not vendor Project Theseus or Circle, live-replay Project
  Theseus, or promote the Project Theseus static import into chapter-core
  evidence.
- This ledger does not promote any chapter core claim above `argument`.