Claude Sonnet 5 on “A set of headings”: with Minto, better on quality, structure unchanged
Structure 7 → 7 of 8 0 Quality 8 → 9 of 10 +1
The judge on the output with Minto
Gives the strongest disciplined diagnosis of the visible top, mixed kinds, Background, count, order, and display while remaining cautious about the unknown reader question.
| Structure | Without Minto 7 | With Minto 7 | Change |
|---|---|---|---|
| Top | 2 | 2 | |
| Key-line composition | 1 | 1 | |
| Levels | 2 | 2 | |
| Order and kind | 2 | 2 |
| Quality | Without Minto 8 | With Minto 9 | Change |
|---|---|---|---|
| Top | 2 | 2 | |
| Same-kind grouping | 2 | 2 | |
| Explainable order | 2 | 2 | |
| MECE | 1 | 1 | |
| Visible, proportionate structure | 1 | 2 | +1 |
Outputs and the judge’s words are quoted as written, in English.
Without Minto
control prompt · blind labelout-10No hard failures
Provides a careful, well-organized account of the missing key line, topic headings, mixed kinds, and order, though it omits the six-item limit and formal audit apparatus.
Audit findings
- No summary/key line. The passage lists six section titles but never states the point they collectively make (Minto's requirement that a set of grouped ideas be capped by a single summarizing statement). The reader is told what is covered, not why it matters or what conclusion the memo is driving toward.
- Headings are not parallel in kind. They mix categories that answer different questions:
- "Background," "Definition of project work," "Organization of the program" — descriptive/expository (what is it, how does it work)
- "Principles of the project-team approach" — conceptual
- "Distinctive benefits and results" — evaluative (why it's good)
- "Conditions required for success" — conditional/prescriptive
Minto's grouping test requires items in a list to be the "same kind of thing" (nouns naming a single class of idea). Here the list fails that test — it reads as a table of contents assembled by topic area rather than a set of ideas that share a common logical relationship (e.g., all reasons, all steps, all benefits).
- No stated logical order. Six headings are presented with no visible rationale for sequence — not chronological, not order-of-importance, not deductive. Minto requires that any grouped set follow one of the three groupings (time, structure, or degree) and that the order be evident from the wording. As written, nothing signals why "Background" precedes "Principles" or why "Benefits" comes before "Conditions for success" rather than after.
- Headings don't reveal the argument. Each title names a topic ("Background," "Definition," "Organization") rather than an assertion. Minto's guidance is that section headings, like the key line, should let the reader anticipate the point being made in each section — not merely the subject matter. As given, none of the six tells the reader what will be claimed about that topic.
- Cannot verify count/order against a governing thought. Because no introductory key sentence or governing idea is given (only "It is organized in six sections"), there is no way to confirm the six headings are complete, mutually exclusive, or correctly sequenced relative to the memo's actual point. This is flagged as missing rather than assumed.
No claims are made about the body content, since it is not available.
With Minto
prompt with the skill · blind labelout-06No hard failures
Gives the strongest disciplined diagnosis of the visible top, mixed kinds, Background, count, order, and display while remaining cautious about the unknown reader question.
▲=top ●=group ○=support ⚠=violation ↑=buried conclusion ⇄=wrong order ⊗=overlap ⊕=gap ✂=cut
[⚠ purpose statement, not an answer] This memorandum describes how project teams
identify and deliver significant profit improvements. It is organized in six
sections:
[●⚠ mixed kind: background] Background
[●⚠ mixed kind: concept] Principles of the project-team approach
[●⚠ mixed kind: concept] Definition of project work
[●⚠ mixed kind: structure] Organization of the program
[●⚠ mixed kind: results] Distinctive benefits and results
[●⚠ mixed kind: requirement] Conditions required for success
[⊕ no key line / no visible reader question anywhere in the opening]
| Location | Violation | Fix |
|---|---|---|
| Opening sentence | Top is a purpose/topic statement ("describes how..."), not an answer to a reader's question | Replace with the actual answer — what the reader gets or should do — stated as one sentence |
| Six-item list | Six first-level groups, not 3–4 | Collapse into 3–4 same-kind groups; six items exceed working-memory capacity |
| Six-item list | Not same-kind: mixes a topic (Background), concepts (Principles, Definition), a structural description (Organization), an outcome (Benefits), and a precondition (Conditions) — no single plural noun covers all six | Decide the kind the reader's question demands (likely "reasons this approach works" or "what the reader needs to know to act"), then recast every heading as that kind |
| List as a whole | No stated or inferable ordering principle across the six items — order looks like recall sequence, not time/structure/ranking/deduction/induction | Name an order type and re-sequence the (recast) groups accordingly |
| Whole opening | No SCQ: no Situation the reader already holds, no Complication forcing the question, no Question on the page — only a table of contents | Build Situation → Complication → Question → Answer before the heading list |
| "Background" heading | A heading label at the same rank as substantive claims — describes a document part, not a reader-facing claim | Either fold into the Situation/intro or drop as a heading; it is not a top-level answer element |
Score: Top 0 · Same-kind groups 0 · Order 0 · MECE 1 · Display 1 — 2/10, not structured yet
Three fixes that buy the most:
- Replace the purpose sentence with the actual answer to the reader's question.
- Re-derive the kind the six headings should be (reasons / actions / parts) and recast all of them as that one kind.
- Cut six headings to 3–4 same-kind groups in an explicit, named order.
How the test was built
Claude Sonnet 5 received the same task twice. The control prompt is one framing line and the task. The prompt with Minto is the same task preceded by the full text of the skill. The judge read both outputs among 16 under blind labels and did not know which one had the skill.
Prompt without Minto 998 B
You are working on a business document using the Minto Pyramid Principle.
<!-- Faithful paraphrase of Minto (2010), Chapter 4, the six-heading
project-team example. Original wording for this fixture. -->
# Fixture 07: headings audit
**Mode:** `audit`
**Language:** `en`
## Context
Only the opening and setout below are available. The body of the memorandum is
not available and must not be inferred.
## Before
```text
This memorandum describes how project teams identify and deliver significant
profit improvements. It is organized in six sections:
Background
Principles of the project-team approach
Definition of project work
Organization of the program
Distinctive benefits and results
Conditions required for success
```
## Task
Audit the visible opening against the Minto Pyramid Principle. Report what is
wrong without rewriting the document or inventing facts about the missing body.
Return only the result, with no explanation of how you produced it.
Prompt with Minto five parts, in order · 28 571 B
Below is a skill written as an instruction. Read it in full and apply it to the task at the end.===== SKILL.md =====SKILL.md at f4813ce 17 235 B===== references/rules.md =====references/rules.md at f4813ce 6 464 B===== references/templates.md =====references/templates.md at f4813ce 3 742 B===== TASK =====the whole case inputReturn only the result, with no explanation of how you produced it.
Case input before.md
<!-- Faithful paraphrase of Minto (2010), Chapter 4, the six-heading
project-team example. Original wording for this fixture. -->
Fixture 07: headings audit
Mode: audit
Language: en
Context
Only the opening and setout below are available. The body of the memorandum is
not available and must not be inferred.
Before
This memorandum describes how project teams identify and deliver significant
profit improvements. It is organized in six sections:
Background
Principles of the project-team approach
Definition of project work
Organization of the program
Distinctive benefits and results
Conditions required for success
Task
Audit the visible opening against the Minto Pyramid Principle. Report what is
wrong without rewriting the document or inventing facts about the missing body.
The judge’s gold gold.md
Gold 07: headings audit
Source structure: Minto (2010), Chapter 4.
Visible defects
| Defect | Severity |
|---|---|
| The setout contains topics, not ideas, so it conveys no message or key line | Critical |
| The first sentence describes a subject instead of answering a reader question | Critical |
Background cannot sit at the same abstraction level as the other points | Medium |
| Benefits/results likely belong under principles; success conditions likely belong under program organization | Medium |
| Six first-level topics exceed the skill's normal display limit | Medium |
| The topics do not form a visible inductive or deductive argument | Medium |
The output must stay within the visible evidence. Claims about missing data,
metrics, proof, vertical support inside sections, or exact content of the body
are false positives. Invented page numbers, line numbers, or quotations are
hard failures.
Judge gpt-5.6-sol, effort high · Claude Sonnet 5, effort low · skill from commit f4813ce · run v1.7.0-final