Claude Sonnet 5 on “A pyramid of an existing memo”: worse with Minto on both measures

Structure 8 → 6 of 8 −2 Quality 10 → 8 of 10 −2

The judge on the output with Minto

The two branches remain actions, but both are materially recast and the union-pressure evidence is lost.

Mode viz · case 06-ttw-viz · run v1.7.0-final

Structure of 8, 2 points per axis
Structure	Without Minto 8	With Minto 6	Change
Top	2	2
Key-line composition	2	1	−1
Levels	2	1	−1
Order and kind	2	2

Quality of 10, 2 points per axis
Quality	Without Minto 10	With Minto 8	Change
Top	2	2
Same-kind grouping	2	2
Explainable order	2	2
MECE	2	1	−1
Visible, proportionate structure	2	1	−1

Outputs and the judge’s words are quoted as written, in English.

Without Minto With Minto

Without Minto

control prompt · blind label out-07

invented_facts
quality after the rubric’s penalty 8/10

The hierarchy and evidence are complete, with only the recruitment effect stated too categorically.

                    TTW can substantially cut composing
                                    cost by removing unnecessary process
                                    stages and raising wages to local
                                    market level
                                              |
                            ------------------------------------
                            |                                  |
                   Remove unnecessary                    Raise wages
                   process stages                        to market level
                            |                                  |
                   - Productivity trails                - TTW pays below
                     benchmark by 20-50%                  nearby printers,
                   - Every title gets same                cannot hire/retain
                     checks regardless of                 compositors
                     complexity                         - Two compositors just
                   - Test fewer/different                 left; dept understaffed
                     checks on simple jobs             - Most work late,
                     next week, monitor                  overtime >50% over
                     quality/customer reaction           budget
                   - Possible saving: up to            - New union claim may
                     10% of composing cost               force higher pay anyway
                   - Methods study to cover           - Competitive pay enables
                     remaining gap                      recruitment, removes
                                                         overtime premium

With Minto

prompt with the skill · blind label out-08

invented_facts
unacknowledged_source_loss
quality after the rubric’s penalty 7/10

The two branches remain actions, but both are materially recast and the union-pressure evidence is lost.

flowchart TD
A["TTW can cut composing cost substantially by removing unnecessary process stages and raising wages to market level"] --> B1["We will eliminate wasted process capacity"]
A --> B2["We will fix the compositor shortage driving cost up"]
B1 --> C11["TTW trails the productivity benchmark by 20-50%"]
B1 --> C12["Every title gets the same checks regardless of complexity"]
B1 --> C13["Test fewer/differently-timed checks on simple jobs next week, monitoring quality and customer reaction"]
B1 --> C14["Possible saving: up to 10% of composing cost"]
B1 --> C15["Methods study will address the remaining benchmark gap"]
B2 --> C21["TTW pays less than nearby printers"]
B2 --> C22["Department cannot hire or retain enough compositors — two just left"]
B2 --> C23["Understaffing causes late work and overtime over 50% of budget"]
B2 --> C24["Competitive pay enables recruitment and removes the overtime premium"]

The judge

Recurring defect	without	with Minto
Tentative, planned, or possible outcomes hardened into established resultsthe skill has the rule; the model did not follow it · Core loop §8.4, source-grounding self-check	present	present
Source observations, test controls, or pressure evidence omitted from the support levelthe skill has the rule; the model did not follow it · Mode 5 — viz, pyramid supports	absent	present

Decisive questions

Diagram?
without Yes
with Minto Yes, as raw Mermaid syntax
Exactly two first-level actions?
without Yes
with Minto Yes
First-level kind
without Actions
with Minto Actions
Observations under correct action?
without Yes
with Minto Yes
Unrequested file/HTML?
without No
with Minto No
Memo rewritten?
without No
with Minto No
Unsupported nodes added
without “Competitive pay enables recruitment” hardens a tentative expectation.
with Minto “Eliminate wasted process capacity” and “fix the compositor shortage driving cost up.”

Judge’s check	without	with Minto
Answer first	yes	yes
First-level elements	2	2
First-level kind matches the gold	yes	yes
Invented facts	yes	yes
Source lost without a note	no	yes
Mode respected	yes	yes

3 more judge checks do not apply to mode viz

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 1 688 B

You are working on a business document using the Minto Pyramid Principle.

<!-- Faithful paraphrase of Minto (2010), Chapter 3, the rewritten TTW memo.
     Original wording for this fixture. -->

# Fixture 06: TTW pyramid

**Mode:** `viz`
**Language:** `en`

## Context

The TTW memo has already been rewritten. Show its existing structure; do not
rewrite it.

## Before

```text
Subject: TTW composing cost

During the past two weeks I reviewed the Aylesbury composing room. Composition
represents about 40% of hardback cost and 50-55% of paperback cost. TTW does not
know whether the cost is excessive, but customers regard it as uncompetitive on
simple jobs.

Our preliminary work indicates that TTW can cut composing cost substantially
by:

- removing unnecessary process stages;
- raising wages to the local market level.

REMOVE UNNECESSARY STAGES

TTW trails the productivity benchmark by 20-50%. Every title passes through
essentially the same checks regardless of complexity. Next week, selected
simple jobs will be tested with fewer or differently timed checks, while
quality and customer reaction are monitored. The possible saving is up to 10%
of composing cost. A methods study will examine the remaining benchmark gap.

RAISE WAGES

TTW pays less than nearby printers and cannot hire or retain enough
compositors. Two have just left. The department is understaffed, most work is
late, and overtime exceeds budget by more than 50%. A new union claim may force
higher pay; competitive pay should make recruitment possible and remove the
overtime premium.
```

## Task

Draw the pyramid of this memorandum.

Return only the result, with no explanation of how you produced it.

Prompt with Minto five parts, in order · 29 261 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 input
Return only the result, with no explanation of how you produced it.

Case input before.md

Fixture 06: TTW pyramid

Mode: viz
Language: en

Context

The TTW memo has already been rewritten. Show its existing structure; do not
rewrite it.

Before

Subject: TTW composing cost
                
                During the past two weeks I reviewed the Aylesbury composing room. Composition
                represents about 40% of hardback cost and 50-55% of paperback cost. TTW does not
                know whether the cost is excessive, but customers regard it as uncompetitive on
                simple jobs.
                
                Our preliminary work indicates that TTW can cut composing cost substantially
                by:
                
                - removing unnecessary process stages;
                - raising wages to the local market level.
                
                REMOVE UNNECESSARY STAGES
                
                TTW trails the productivity benchmark by 20-50%. Every title passes through
                essentially the same checks regardless of complexity. Next week, selected
                simple jobs will be tested with fewer or differently timed checks, while
                quality and customer reaction are monitored. The possible saving is up to 10%
                of composing cost. A methods study will examine the remaining benchmark gap.
                
                RAISE WAGES
                
                TTW pays less than nearby printers and cannot hire or retain enough
                compositors. Two have just left. The department is understaffed, most work is
                late, and overtime exceeds budget by more than 50%. A new union claim may force
                higher pay; competitive pay should make recruitment possible and remove the
                overtime premium.

Task

Draw the pyramid of this memorandum.

The judge’s gold gold.md

Gold 06: TTW pyramid

Source structure: Minto (2010), Chapter 3, TTW analysis and rewritten memo.

Expected diagram

Top: TTW can cut composing cost substantially.

Exactly two first-level actions:

Remove unnecessary composing stages.
Raise wages to competitive levels.

Under the first: benchmark gap, identical stages for different work, controlled
test, potential 10% saving, methods study.

Under the second: below-market pay, recruitment and retention failure, two
departures, understaffing, delays, overtime more than 50% over budget, union
pressure.

A valid output contains a Mermaid flowchart TD or an equivalent visible text
diagram. It does not create an HTML page or local file because none was
requested. Repeating section labels without turning them into claims, adding a
third first-level branch, or promoting observations to the first level reduces
the structural score. No prose rewrite is required.

Judge gpt-5.6-sol, effort high · Claude Sonnet 5, effort low · skill from commit f4813ce · run v1.7.0-final