# Reader Continuity Audit

Status: generated heuristic report for Phase 2 reader review.

This report is generated by `python3 scripts/audit_reader_continuity.py --write` from a temporary reader-edition workspace. It is not a manual full-book review, not a release record, and not proof that the normal reader manuscript is publication-ready.

## Summary

| Metric | Value |
|---|---:|
| Reader chapters measured | 87 |
| Generated files | 92 |
| Live-only sections removed before audit | 432 |
| Human-only bridges unwrapped | 142 |
| Reader scaffold terms humanized | 54 |
| Active reader overlay operations | 75 |
| Applied reader overlay operations | 75 |
| Reader words measured | 439,166 |
| Minimum chapter words | 2,724 |
| Median chapter words | 4,711 |
| Maximum chapter words | 10,364 |
| High-priority heuristic review chapters | 67 |
| Medium-priority heuristic review chapters | 12 |
| Low-priority heuristic review chapters | 8 |
| Table rows in generated reader chapters | 1094 |
| Mermaid diagrams in generated reader chapters | 116 |
| Non-Mermaid code blocks in generated reader chapters | 18 |
| Paragraphs at or above 160 words | 25 |
| Repeated first-sentence stems detected | 0 |

## How To Use This Audit

- Treat high-priority rows as the first candidates for human continuity review, not as defects.
- Classify each real issue as a canonical live-book edit, a reader-only overlay, companion-note treatment, or no action.
- Preserve claim text, support boundaries, implementation horizons, and non-claim language inherited from the live book.
- Do not render or publish reader artifacts from this audit alone.

## Review Queue

| Priority | Score | Chapter | Words | Dense hits | Tables | Code | Long paragraphs | Reason |
|---|---:|---|---:|---:|---:|---:|---:|---|
| high | 165 | `human-factors-and-meaningful-control-in-oversight` | 4,486 | 33 | 80 | 0 | 0 | long; 80 table row(s) |
| high | 144 | `governed-world-models-and-reality-grounding` | 5,985 | 55 | 68 | 0 | 0 | very long; 68 table row(s) |
| high | 138 | `governed-operations-incident-command-and-graceful-degradation` | 7,227 | 51 | 65 | 0 | 0 | very long; 65 table row(s) |
| high | 130 | `learning-compute-topology-and-adaptive-process-architecture` | 5,998 | 41 | 62 | 0 | 0 | very long; 62 table row(s) |
| high | 110 | `rankfold-neuralfold-and-artifact-compression` | 6,899 | 72 | 50 | 0 | 0 | very long; 50 table row(s) |
| high | 106 | `replaceable-cognitive-substrates-beyond-transformer-monoculture` | 7,936 | 65 | 47 | 2 | 1 | very long; 1 long paragraph(s); 47 table row(s); 2 code block(s) |
| high | 106 | `governed-model-training-distributed-optimization-and-scaling` | 6,926 | 63 | 48 | 2 | 0 | very long; 48 table row(s); 2 code block(s) |
| high | 106 | `white-box-evidence-interpretability-and-activation-governance` | 5,072 | 29 | 50 | 0 | 0 | very long; 50 table row(s) |
| high | 77 | `fast-generation-architectures` | 6,931 | 76 | 31 | 0 | 2 | very long; 2 long paragraph(s); 31 table row(s) |
| high | 63 | `ai-work-surfaces-agent-harnesses-and-organizational-absorption` | 4,110 | 23 | 30 | 0 | 0 | long; 30 table row(s) |
| high | 62 | `compact-generative-systems-and-residual-honesty` | 8,947 | 110 | 23 | 0 | 1 | very long; 1 long paragraph(s); 23 table row(s) |
| high | 61 | `security-kernel-and-digital-scifs` | 5,486 | 55 | 23 | 0 | 2 | very long; 2 long paragraph(s); 23 table row(s) |

## Repeated Opening Stems

No repeated first-sentence stems were detected with the current eight-word heuristic.

## Chapter Metrics

| Chapter | Words | Paragraphs | Dense hits | Tables | Diagrams | Code | Long paragraphs | Priority |
|---|---:|---:|---:|---:|---:|---:|---:|---|
| `asi-is-a-stack-not-a-model` | 4,182 | 59 | 55 | 5 | 2 | 0 | 0 | high |
| `the-efficient-asi-hypothesis` | 4,436 | 57 | 66 | 7 | 1 | 0 | 0 | high |
| `system-boundaries-and-authority` | 3,827 | 57 | 81 | 0 | 3 | 0 | 0 | medium |
| `failure-modes-of-ungoverned-intelligence` | 4,894 | 69 | 63 | 6 | 1 | 0 | 0 | high |
| `dangerous-capability-domains-and-misuse-uplift` | 4,338 | 80 | 20 | 8 | 1 | 0 | 0 | high |
| `military-ai-autonomous-weapons-and-strategic-stability` | 4,130 | 72 | 24 | 10 | 1 | 1 | 0 | high |
| `evidence-states-and-claim-discipline` | 4,666 | 66 | 155 | 1 | 2 | 0 | 0 | high |
| `scalable-oversight-and-adversarial-ai-control` | 5,485 | 72 | 56 | 5 | 1 | 0 | 0 | high |
| `human-intent-as-a-formal-input` | 4,727 | 67 | 47 | 0 | 1 | 0 | 0 | medium |
| `human-factors-and-meaningful-control-in-oversight` | 4,486 | 61 | 33 | 80 | 2 | 0 | 0 | high |
| `human-ai-communication-persuasion-and-epistemic-security` | 3,839 | 76 | 25 | 0 | 1 | 0 | 0 | low |
| `constitutional-alignment-substrate` | 4,203 | 48 | 55 | 0 | 1 | 0 | 1 | medium |
| `inner-alignment-mesa-optimization-and-learned-objective-integrity` | 5,258 | 85 | 33 | 15 | 2 | 1 | 0 | high |
| `moral-uncertainty-and-value-conflict` | 4,711 | 49 | 86 | 0 | 1 | 0 | 2 | high |
| `governed-objective-formation-value-learning-and-goal-integrity` | 3,622 | 74 | 24 | 0 | 1 | 0 | 0 | low |
| `institutions-international-coordination-and-public-legitimacy` | 4,129 | 82 | 29 | 0 | 1 | 0 | 0 | low |
| `societal-resilience-and-misuse-defense` | 4,656 | 81 | 27 | 12 | 1 | 0 | 0 | high |
| `stable-capability-fields` | 4,941 | 67 | 79 | 0 | 1 | 0 | 0 | medium |
| `capability-replacement-and-rollback` | 4,613 | 63 | 76 | 3 | 1 | 0 | 0 | high |
| `security-kernel-and-digital-scifs` | 5,486 | 69 | 55 | 23 | 2 | 0 | 2 | high |
| `adversarial-machine-learning-and-model-attack-surface` | 3,991 | 86 | 52 | 5 | 1 | 0 | 0 | high |
| `privacy-data-rights-and-information-flow-governance` | 4,204 | 52 | 62 | 7 | 1 | 1 | 1 | high |
| `confidential-and-verifiable-ai-computation` | 3,817 | 46 | 45 | 9 | 1 | 1 | 0 | high |
| `model-weight-custody-and-hardware-roots-of-trust` | 7,338 | 88 | 89 | 21 | 1 | 0 | 1 | high |
| `open-weight-release-and-post-release-control` | 3,859 | 75 | 30 | 19 | 1 | 0 | 0 | high |
| `ai-supply-chain-integrity-and-lifecycle-provenance` | 5,705 | 69 | 88 | 13 | 1 | 0 | 2 | high |
| `recursive-self-improvement-boundaries` | 4,026 | 63 | 65 | 0 | 2 | 0 | 0 | medium |
| `open-ended-improvement-engines` | 5,078 | 72 | 62 | 7 | 1 | 0 | 1 | high |
| `autonomous-replication-proliferation-and-containment` | 3,997 | 86 | 28 | 6 | 1 | 0 | 0 | high |
| `intent-to-execution-contracts` | 5,145 | 73 | 71 | 0 | 2 | 0 | 0 | high |
| `perception-sensor-fusion-and-observation-trust` | 3,977 | 55 | 44 | 15 | 2 | 1 | 0 | high |
| `planning-as-a-control-layer` | 4,666 | 72 | 49 | 0 | 1 | 0 | 0 | low |
| `governed-world-models-and-reality-grounding` | 5,985 | 81 | 55 | 68 | 2 | 0 | 0 | high |
| `cognitive-compilation-and-semantic-ir` | 7,156 | 94 | 166 | 20 | 1 | 1 | 0 | high |
| `virtual-context-abi` | 6,866 | 93 | 103 | 15 | 1 | 0 | 0 | high |
| `durable-semantic-memory-and-knowledge-lattices` | 5,360 | 96 | 60 | 0 | 1 | 0 | 0 | medium |
| `context-transactions-snapshots-mounts-and-taint` | 5,555 | 63 | 94 | 0 | 1 | 0 | 0 | medium |
| `verification-bandwidth-and-context-adequacy` | 4,720 | 53 | 75 | 0 | 1 | 0 | 0 | medium |
| `claim-ledgers-and-belief-revision` | 5,243 | 57 | 138 | 13 | 1 | 0 | 1 | high |
| `spinoza-verification-and-proof-carrying-claims` | 4,838 | 59 | 140 | 7 | 1 | 0 | 0 | high |
| `labor-os-and-typed-jobs` | 4,461 | 53 | 81 | 9 | 1 | 0 | 0 | high |
| `ai-work-surfaces-agent-harnesses-and-organizational-absorption` | 4,110 | 61 | 23 | 30 | 1 | 0 | 0 | high |
| `human-ai-organizations-delegation-and-accountability` | 5,374 | 94 | 35 | 7 | 2 | 1 | 0 | high |
| `human-ai-symbiosis-neurotechnology-and-cognitive-sovereignty` | 3,845 | 51 | 15 | 8 | 1 | 0 | 0 | high |
| `ai-deployment-transition-distribution-and-human-agency` | 3,700 | 78 | 18 | 0 | 1 | 0 | 0 | low |
| `artifact-graphs-audit-logs-and-replay` | 4,974 | 62 | 155 | 1 | 1 | 0 | 0 | high |
| `runtime-adapters-tool-permissions-and-human-approval` | 6,350 | 78 | 100 | 0 | 2 | 0 | 0 | high |
| `embodied-agency-real-time-control-and-physical-safety` | 4,076 | 60 | 23 | 14 | 2 | 1 | 0 | high |
| `inter-stack-protocols-identity-and-economic-exchange` | 5,106 | 60 | 76 | 7 | 1 | 0 | 0 | high |
| `multi-agent-dynamics-collective-intelligence-and-systemic-risk` | 4,663 | 87 | 32 | 7 | 2 | 1 | 0 | high |
| `procedural-memory-and-cognitive-loop-closure` | 6,651 | 93 | 83 | 8 | 1 | 0 | 0 | high |
| `routing-heads-and-specialist-cores` | 7,498 | 103 | 132 | 8 | 2 | 1 | 0 | high |
| `replaceable-cognitive-substrates-beyond-transformer-monoculture` | 7,936 | 103 | 65 | 47 | 1 | 2 | 1 | high |
| `relational-dimension-compilation-and-polyadic-cognition` | 4,306 | 55 | 48 | 12 | 1 | 1 | 0 | high |
| `governed-model-training-distributed-optimization-and-scaling` | 6,926 | 90 | 63 | 48 | 3 | 2 | 0 | high |
| `learning-compute-topology-and-adaptive-process-architecture` | 5,998 | 87 | 41 | 62 | 1 | 0 | 0 | high |
| `learning-theory-generalization-and-scaling-science` | 4,252 | 80 | 36 | 10 | 1 | 0 | 0 | high |
| `readiness-gates-residual-escrow-and-quarantine` | 5,899 | 80 | 105 | 11 | 2 | 0 | 0 | high |
| `personal-compute-hives-and-federated-edge-intelligence` | 5,634 | 71 | 66 | 6 | 1 | 0 | 1 | high |
| `compact-generative-systems-and-residual-honesty` | 8,947 | 120 | 110 | 23 | 3 | 0 | 1 | high |
| `fast-generation-architectures` | 6,931 | 87 | 76 | 31 | 3 | 0 | 2 | high |
| `governed-deliberation-and-test-time-scaling` | 5,372 | 64 | 41 | 21 | 1 | 0 | 0 | high |
| `rankfold-neuralfold-and-artifact-compression` | 6,899 | 108 | 72 | 50 | 2 | 0 | 0 | high |
| `resource-economics-and-token-budgets` | 10,364 | 142 | 122 | 5 | 2 | 2 | 2 | high |
| `physical-compute-infrastructure-energy-and-environmental-constraints` | 3,891 | 81 | 20 | 0 | 1 | 0 | 0 | low |
| `mathematical-and-search-substrates` | 4,829 | 62 | 95 | 0 | 1 | 0 | 1 | medium |
| `circle-calculus-and-proof-carrying-ai-contracts` | 2,724 | 43 | 107 | 0 | 1 | 0 | 0 | medium |
| `coil-attention-cyclic-memory-and-recurrence-contracts` | 3,445 | 55 | 54 | 0 | 1 | 0 | 0 | low |
| `coilra-multicoil-rope-and-cyclic-mixers` | 4,551 | 68 | 158 | 0 | 1 | 0 | 0 | medium |
| `executable-specifications-and-lean-proof-envelope` | 2,794 | 51 | 125 | 0 | 1 | 0 | 0 | medium |
| `benchmark-ratchets-and-anti-goodhart-evidence` | 7,585 | 108 | 168 | 21 | 2 | 0 | 0 | high |
| `white-box-evidence-interpretability-and-activation-governance` | 5,072 | 71 | 29 | 50 | 2 | 0 | 0 | high |
| `capability-thresholds-and-deployment-commitments` | 4,679 | 67 | 54 | 7 | 1 | 0 | 1 | high |
| `adversarial-evaluation-sandbagging-and-training-time-deception` | 5,309 | 79 | 61 | 7 | 1 | 0 | 0 | high |
| `safety-cases-and-structured-assurance` | 4,233 | 60 | 76 | 13 | 1 | 0 | 0 | high |
| `content-authenticity-watermarking-and-synthetic-media-integrity` | 4,446 | 82 | 52 | 9 | 1 | 0 | 0 | high |
| `governed-operations-incident-command-and-graceful-degradation` | 7,227 | 97 | 51 | 65 | 1 | 0 | 0 | high |
| `adjudicated-persistence-and-the-adaptive-commit-boundary` | 4,511 | 67 | 35 | 16 | 1 | 0 | 0 | high |
| `policy-optimization-and-learning-from-feedback` | 5,138 | 71 | 81 | 1 | 2 | 0 | 1 | high |
| `data-engines-continual-learning-and-unlearning` | 6,989 | 96 | 109 | 21 | 2 | 0 | 1 | high |
| `scientific-discovery-and-experimental-governance` | 4,433 | 83 | 32 | 0 | 1 | 0 | 0 | low |
| `artifact-steward-agents-and-living-project-governance` | 5,378 | 87 | 100 | 0 | 1 | 0 | 1 | high |
| `integrated-reference-architecture` | 5,545 | 73 | 137 | 2 | 1 | 0 | 2 | high |
| `project-theseus-as-report-first-implementation-reference` | 3,963 | 55 | 88 | 16 | 1 | 1 | 0 | high |
| `prototype-roadmap` | 3,249 | 48 | 66 | 8 | 1 | 0 | 0 | high |
| `living-book-methodology` | 3,907 | 61 | 120 | 14 | 1 | 0 | 0 | high |
| `open-research-agenda-and-bibliography-plan` | 2,841 | 47 | 68 | 19 | 1 | 0 | 0 | high |

## Non-Claims

- This audit does not claim a reviewed reader release exists.
- This audit does not claim EPUB, PDF, DOCX, audio, or audio-embedded EPUB artifacts exist.
- This audit does not promote any claim support state.
- This audit does not prove reader quality; it only creates a deterministic queue for human review.