| ISBN | Format | Status |
|---|---|---|
| 979-8-9964101-0-1 | eBook (ePub) KDP · D2D · Apple Books · Kobo · B&N | Assigned |
| 979-8-9964101-1-8 | Hardcover (Print) KDP Print or IngramSpark hardcover edition | Assigned |
| 979-8-9964101-2-5 | Paperback (Print) Primary print edition — back cover ISBN barcode | Assigned |
| 979-8-9964101-3-2 | Audiobook Reserved — not yet in production | Reserved |
| 979-8-9964101-4-9 through -9-4 | Reserve Pool (6 remaining) Unassigned — available for future editions | Unassigned |
Margins are estimates. Verify at kdp.amazon.com and myaccount.ingramspark.com before setting final prices.
| Ch | Title | Part |
|---|---|---|
| Front Matter | ||
| — | Preface | |
| — | Introduction | |
| Part One — The Phenomena | ||
| 1 | Impossible Things Before Breakfast | Part One — Phenomena |
| 2 | On Radar, On Camera, On Record | Part One — Phenomena |
| 3 | Off the Books | Part One — Phenomena |
| 4 | What We Agreed to See | Part One — Phenomena |
| Part Two — The Architecture of Secrecy | ||
| 5 | How Secrets Are Made | Part Two — Secrecy |
| 6 | Special Access | Part Two — Secrecy |
| 7 | What NASA Saw | Part Two — Secrecy |
| 8 | Standing Orders | Part Two — Secrecy |
| 9 | Manufacturing Doubt | Part Two — Secrecy |
| Part Three — The Evidence | ||
| 10 | Sworn | Part Three — Evidence |
| 11 | On the Evidence | Part Three — Evidence |
| 12 | Theories of Intent | Part Three — Evidence |
| 13 | The Distance Problem | Part Three — Evidence |
| 14 | Science Rushes In | Part Three — Evidence |
| 15 | Controlled Disclosure or Ontological Shock? | Part Three — Evidence |
| Part Four — Disclosure | ||
| 16 | Congressional Lockout | Part Four — Disclosure |
| 17 | The President Doesn’t Have Clearance | Part Four — Disclosure |
| 18 | The Public Decides | Part Four — Disclosure |
| Back Matter | ||
| — | Conclusion | |
| — | Appendices A1–A15 | |
| — | Further Reading | |
| — | Bibliography | |
| — | Glossary | |
| — | Index | |
| — | About the Author | |
| — | Acknowledgements | |
Trigger passes by typing the command phrase to Claude with the master uploaded. Scope is set by the command — Claude will report what is already logged clean versus open, and propose a reasonable chunk if the volume is large. Logs are written before every pack.
Triggered by prompt scope, not document structure. Run before any Tier B work on the same content. All eight sub-passes run together unless you call one individually.
| Command | Sub-pass | What it covers |
|---|---|---|
| "Prose pass [scope]" | Full Tier A | All eight sub-passes on the stated scope (e.g. Ch 5, Chs 1–4, book-wide). |
| "Word-by-word [scope]" | A1 | Duplicate words, comma splices, doubled punctuation, missing words. |
| "Spelling check [scope]" | A2 | Spelling, grammar, capitalization, smart quotes, em-dash audit, institutional singular, Over-classified, Government. |
| "AI-speak [scope]" | A3 | Watch-list scan. Choice form per hit: Keep / Use proposed / Write my own. |
| "Paragraph logic [scope]" | A4 | Each paragraph opens, develops one idea, closes cleanly. Flags over-150-word paragraphs and parallel openers. |
| "Paragraph structure [scope]" | A5 | Too-short paragraphs (≤2 sentences, can't stand alone); over-long without natural break. |
| "Flow check [scope]" | A6 | Prohibited patterns: em-dashes in body, throat-clearing, Not-X constructions, passive voice, bullet lists. Voice, register, rhythm. |
| "Thematic echo [scope]" | A7 | Repeated images, phrases, or structural moves within a chapter or across adjacent chapters. |
| "Sterling [scope]" | A8 | Chapter title weight, pull quote differentiation, connective tissue, tonal register. Advisory notes unless significant issue found. |
Run after Tier A is confirmed clean for the same content. Each sub-pass is discrete and writes a timestamped log entry before packing. If content is edited after a log entry, that passage is flagged for re-check at next session open.
| Command | Sub-pass | What it covers |
|---|---|---|
| "Formatting pass [scope]" | B1 | Style ID compliance, image placement rule, caption construction, section breaks, pull quote/Attribution pairs, TextOpen after subheadings, chapter structure order, two-space title normalization. |
| "Citation pass [scope]" | B2 | Four-point check: (1) every significant factual claim has a marker; (2) all citations are primary sources per hierarchy; (3) each citation is fetched and verified to support the specific claim; (4) every citation appears in A15 Bibliography with a complete entry. Logs to Citation_Verification_Log.md. |
| "Cross-ref pass [scope]" | B3 | Every "see Chapter N" and "Appendix AX" reference verified to resolve to real content with matching material. Back matter included unless Dave specifies otherwise. Logs to CrossRef_Verification_Log.md. |
| "Footnote pass [scope]" | B4 | Four-count gate (ref = note bookmarks = forward = return hyperlinks). Adjacent marker audit. Sequence integrity 1→N. afootnote construction check. No orphaned markers. |
| "Appendix pass" | B5 | A2 sightings scan, A12 figures/diagrams, A15 Quotations location codes, A15 Bibliography new entries, Glossary new terms, Index new subjects. Run after B2 is confirmed for relevant chapters. |
| "URL pass" | B6 | Every hyperlink verified to resolve. Invisible-hyperlink rendering confirmed (color=auto, u=none, no Hyperlink style). |
| "Index pass" | B7 | Run last — after all content is locked. Locator codes, bookmark naming, subentry formatting, print vs. Kindle locator styles, exclusion rules (A15, A12). Editing body content after B7 potentially invalidates locators. |
Two log files are maintained as project files. Entries are written at the moment work is done — never reconstructed. Logs start clean; prior v228/229 work is not backfilled.
| File | What it records |
|---|---|
| Citation_Verification_Log.md | One entry per B2 pass: chapter(s), version, date, citations checked, source fetch result (confirmed / flagged / unreachable), agreed exceptions. |
| CrossRef_Verification_Log.md | One entry per B3 pass: scope, version, date, references verified, mismatches found and resolved. |
Run only after Tier A and B are confirmed clean for all content. C0 precedes C1 and C2.
| Command | Pass | What it covers | Status |
|---|---|---|---|
| "Final darlings review" | C0 | One full read for anything that slipped through. Minor prose only — no structural changes. Sterling final commercial assessment. | |
| "Production pass ebook" | C1 | Heading hierarchy, TOC integrity, front/back matter, metadata, cross-reference integrity, image resolution (72–150 DPI), table simplification, manual line break removal, hyperlink audit. | Done v230 |
| "Production pass print" | C2 | Image resolution (300 DPI), color/grayscale audit, margin/gutter (6″×9″, 1″ margins), page breaks on recto, widows/orphans, index locator style (sz=16 print override), print index completeness. | Next priority |
| When | What Claude does |
|---|---|
| Session open | Reports in one line: what is logged clean, what is open, and what content has been touched since the last verified pass. |
| Session close | Presents continuation prompt. Reminds Dave to attach the hub JSON export at the start of the next session so any new hub items are picked up. |
| Before every pack | Writes log entry for any Tier B pass completed this session. Runs Pre-Pack Gate (sectPr integrity, style compliance, para structure, four-count). |
The Roadmap is a Gantt-style view of your project timeline. It organizes work into phases, each containing tasks and optional gates (decision or approval checkpoints).
The Notes tab is the main working surface. Items can be standalone captures, reminders, questions, or two-way dialogue threads with Claude. Everything here is included in the JSON export Claude reads at the start of each session.
Expand any note and tap → Roadmap to promote it to a Roadmap phase or task. The note remains in Notes as a reference.
blue = awaiting Claude. amber = awaiting Dave. When Claude’s reply is the latest entry, a reply field appears inline so you can continue the thread. All thread data is included in the JSON export.
Resources is a curated link library organized by category — primary sources, substacks, news feeds, analysis, podcasts, documentaries, and project references. Use it to track sources you return to regularly.
Tap the pencil icon on any source row to edit it. Tap the trash icon to delete (deletion is immediate).
Outreach tracks contacts for blurb requests, pitch submissions, and other external outreach. Confirmed, closed, and rejected contacts are hidden by default — use the filter to show them.
Tap + Add Contact. Enter name, role, outreach type (blurb / pitch / other), status, and any notes. A contact can carry a URL (author website, publisher page, etc.).
Expand a contact and tap Edit to update status, add follow-up notes, or record a response date. Moving a contact to Confirmed or Closed hides it from the default view.
Reference is a read-only quick-access deck. It does not store items you create — it contains fixed reference material Claude and Dave use during working sessions.
Hover over or tap Reference ▸ in the main tab bar to open the flyout menu. Select a section. The content loads inline below the tab bar.
The hub is a single HTML file stored in a private GitHub repository. GitHub Pages serves it as a live website accessible from any browser on any device. There is no server, no database, and no account login — the file itself is the application.
Your notes, roadmap, and outreach data live in two places simultaneously: in localStorage on the device where you made changes, and in the GitHub file after a sync. localStorage is a per-device buffer — it does not automatically travel to other devices. The GitHub file is the single source of truth.
When you tap Update Claude, the hub copies a JSON snapshot of your current data to the clipboard. You paste that into Claude. Claude reads the snapshot, fetches the live file directly from GitHub, applies your changes surgically, and pushes the updated file back to GitHub. Refreshing the hub on any device after that gives you the latest version.
Collision prevention
Because the GitHub file uses last-write-wins, avoid running two Claude sessions that both push to GitHub at the same time. Claude always fetches the live file before making changes — but if two sessions push simultaneously, one will overwrite the other. One session writing at a time is the safe pattern.
Used in the Update Claude modal. Always paste into this project — not a new chat.
Stored locally on this device only — never in the JSON export.
When you click ↻ Refresh in Resources, only sources tagged with checked keywords have their Last Checked timestamp updated.
Shared tag list used across Notes, Resources, and Outreach. Add by chapter, part, topic, or anything you reference repeatedly.
The Update Claude modal shows each time unless you checked “Don’t show this again.” Click below to restore it.
Hub v6 · brownd11-a11y.github.io/oc-hub