---
review: r001
book_version: v0.1.0
reviewed: 2026-08-28
reviewer: the founder
source: v2/briefs/40__founder-memo__the-agentic-team-and-who-the-book-is-for.md
state: actioned
---

# r001 — the title does not hold, and the reading that says so was not on the record

The founder read the book at **v0.1.0** and opened brief 40 with the finding:

> *"one of the first interesting questions to ask is, is the name of the book correct?
> And I don't think it is."*

That reading produced the retitle at v0.6.9 and the change-control pack that carried it.
**It was never recorded as a review**, which is why this file exists and is dated after
the work it describes. The estate has had a review register since v0.3.7 and did not use
it for the one substantive reading of this book.

Recording it late is worth doing anyway. A pack records a decision; a review records what
a reader saw, including the parts the decision did not act on.

## Item 1 — the title names the two rarest things in the book
**State:** actioned
**Outcome:** retitled to *Creating a Book Using Agentic Workflows* at v0.6.9, book v0.2.0.

Counted rather than argued, in [the researcher's map](../../../team/researcher/debriefs/making-a-book__v0.1.0__map-the-making-of-book.md):
*fractal* appeared **9 times in 31,221 words**, six of them quoting the other book's
title, and exactly once about anything this book does. *Semantic graph* appeared 6 times.
Against *release* 227, *graph* 193, *book* 182, *agent* 173.

The book had already said so about itself: its front matter noted the title *"was set in
the commission"*, unlike everything else in it.

## Item 2 — the new title is not earned by the body either
**State:** open
**Outcome:** open. Tracked as `@editor/issues/open/001` and WS-02 on the board.

Recorded on the day the retitle shipped rather than discovered later. The book's own
graph counts *workflow* and *workflows* **6 times in 27,002 words** of prose, and the
vocabulary the founder uses for the idea — *villager*, *zone*, *pacing*, *maturity*,
*commoditised* — appears **zero** times.

The title becomes true when part one and the vocabulary pass land. Shipping the rename
first was deliberate; leaving this unstated would not have been.

## Item 3 — does the cover name Claude, or agents?
**State:** open
**Outcome:** open, and blocked on the founder. `@editor/issues/blocked/002`.

Brief 40 said *"writing a book with Claude"*. Brief 41 said *"making a book with agents"*.
The book names Claude **4 times in 27,002 words**. The two memos point different ways and
the question has been open since v0.6.3.

## Item 4 — the book leans villager, and brief 40 read it as explorer
**State:** superseded
**Outcome:** superseded by brief 45. Neither reading was right.

The map found the body carries **one code block and no shell commands** across twelve
chapters, with the technical material quarantined in an appendix that is 21% of the words:
villager-shaped as written. Brief 40 judged the book explorer-facing. That was recorded as
a contradiction for the founder to settle.

Brief 45 settles it and dissolves it: *"sometimes the difference between somebody with more
or less experience is just that they consume materials faster, but they still go to the same
steps."* Villager and explorer are phases the same person moves between. The book being
villager-shaped and an explorer reading it faster are the same fact, so there was never a
contradiction to settle.

## Item 5 — the colophon claimed the second book was unwritten
**State:** actioned
**Outcome:** corrected at v0.6.9, in the same pass as the retitle.

True at v0.5.11, false by the time anyone read it. The colophon now scopes the claim to
when it was true and says what happened since, rather than being edited to look as though
it had never been wrong.