qwen3.6-fp8 on “A set of monthly reports”: with Minto, better on quality, structure unchanged

Structure 4 → 4 of 8 0 Quality 5 → 7 of 10 +2

The judge on the output with Minto

It is visually disciplined but mistakes the buried streamlining direction for the recommendation management requested.

Mode audit · case 03-period-graph-books · run v1.7.0-final

Structure of 8, 2 points per axis
Structure	Without Minto 4	With Minto 4
Top	1	1
Key-line composition	1	1
Levels	1	1
Order and kind	1	1

Quality of 10, 2 points per axis
Quality	Without Minto 5	With Minto 7	Change
Top	1	1
Same-kind grouping	2	2
Explainable order	1	1
MECE	0	1	+1
Visible, proportionate structure	1	2	+1

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

Without Minto With Minto

Without Minto

control prompt · blind label out-16

invented_facts

It remains concise and diagnosis-only, but treats streamlining as the buried recommendation and adds fabricated urgency and timing.

Minto Pyramid Principle Audit: Period Graph Books

Defect: Conclusion and recommendation are buried at the end. The memo opens with historical background and procedural steps, delaying the core message until the "Overall assessment" section.
Required Change: State the main conclusion and recommended action at the very beginning. Lead with a single, decisive sentence that answers the implicit management question (e.g., "The graph book production process requires immediate streamlining to eliminate recurring errors and clarify accountability.").
Defect: Lack of top-down hierarchical structure. Supporting details (production steps, specific errors, process flaws) are presented in a flat, sequential narrative rather than organized under a clear umbrella statement.
Required Change: Restructure so that every subsequent section explicitly supports the opening conclusion. Group related findings under 3–5 distinct, logically parallel headings that collectively justify the need for change.
Defect: Failure to apply MECE (Mutually Exclusive, Collectively Exhaustive) grouping. The "Issues" section mixes symptoms (incorrect calculations, inconsistent data), process flaws (multiple data entries), and organizational gaps (unclear responsibility) without categorical separation.
Required Change: Categorize supporting arguments into distinct, non-overlapping dimensions (e.g., Process Design, Data Integrity, Accountability). Ensure each category contains all relevant evidence needed to support the main conclusion.
Defect: Absence of SCQA framing and logical sequencing. The document jumps directly into operational steps without establishing the Situation, Complication, and Key Question that naturally drive the analysis.
Required Change: Introduce a brief SCQA context upfront: establish the baseline (Situation), identify the disruption caused by the transfer (Complication), pose the resulting management question, and immediately follow with the Answer/Recommendation. Order supporting points logically (e.g., by impact severity or process phase).
Defect: Weak, non-actionable closing. The memo ends with an open-ended discussion request rather than a clear decision point or proposed next step.
Required Change: Conclude with a specific, time-bound recommendation or decision requested from senior management (e.g., "Approve a cross-functional task force to redesign the data-entry workflow within 30 days."). Ensure the closing directly mirrors the opening recommendation to maintain pyramid symmetry.

With Minto

prompt with the skill · blind label out-07

No hard failures

It is visually disciplined but mistakes the buried streamlining direction for the recommendation management requested.

▲ ⚠ ● ○ ⚠ ↑ ⇄ ⊗ ⊕ ✂

Marker	Meaning
`▲`	top / answer
`●`	first-level group
`○`	supporting detail
`⚠`	violation (mixed kinds, unexplainable order, false grouping)
`↑`	conclusion buried lower than it belongs — promote it
`⇄`	wrong order type here
`⊗`	overlap (ME broken)
`⊕`	gap (CE broken)
`✂`	cut, or move to an appendix

⚠ Subject: Period Graph Books
                
                Since Period 5, Financial Analysis has been responsible for producing four
                graph-presentation books previously handled by Corporate Planning. This memo
                describes issues that appeared after the transfer. ✂
                
                ● ⚠ Production
                1. Gather data from divisional forms, internal documents, and telephone calls.
                2. Calculate graph points manually or on the corporate computer, including
                   rolling values for revenue, cost, and percentages.
                3. Transcribe the points to input sheets. Analysts update the current period,
                   then Color Graphics re-enters the points into its own database and generates
                   the graphs.
                4. Check the data and calculations.
                
                ● ⚠ Issues
                The transfer inserted another person into an already fragmented process.
                Divisional staff and three corporate departments participate, and the same
                information may be entered into computer systems three times.
                
                For one book, divisional data are loaded into a dedicated corporate program,
                calculated, transferred to Color Graphics input sheets, entered again, and
                returned for updates. Problems include inconsistent data between periods and
                regions, incorrect variable-cost calculations, unexplained changes to points
                that were previously correct, and prior-period data that must be re-entered.
                
                ↑ ⚠ Overall assessment
                Most problems arise from a cumbersome fragmented process. ⊕ No one controls the
                data from beginning to end, responsibility is unclear, and errors can fall
                between stages. The production process badly needs streamlining, both for each
                book and for common calculations. The current team cannot manage the required
                changes as well as ongoing production.
                
                ⚠ Could we discuss how best to proceed?

Findings

Location	Violation	Fix
First paragraph / Subject	Top does not answer; topic label	Promote "Overall assessment" to top; state recommendation to streamline process.
"Overall assessment"	Buried logic; Vertical gap	Move conclusion to top; move "No one controls" to supports under a group.
"Production"	Wrong group kind; Context disguised as argument	Demote to intro context or supports under "Fragmentation causes errors"; remove as top-level group.
"Issues"	Mixed kinds; Evidence without grouping	Group elements as reasons for errors or symptoms; ensure same-kind classification.
Intro	Meta-commentary	Remove "This memo describes..."; draft SCQ story dosed to reader's knowledge.
Closing	Vague ask	Request specific decision on streamlining proposal rather than open discussion.

Score

Axis	Score
Top	0
Same-kind groups	0
Order	0
MECE	0
Display	1
Total	1

Band: 0–3 not structured yet

Three fixes that buy the most

Move the recommendation to the top: "We recommend streamlining the graph book process to eliminate errors and clarify data control."
Rebuild groups to support the top: e.g., 1. Fragmentation causes recurring errors and inconsistencies. 2. Current team lacks capacity to redesign the process while maintaining output.
Replace the intro with an SCQ story dosed to the reader's knowledge of the transfer and the resulting errors.

The judge

Recurring defect	without	with Minto
Missing answer is not distinguished from the buried streamlining conclusionthe skill has the rule; the model did not follow it · Mode 2 — audit, check 1: Top	present	present
Incomplete process logic is not separated into graph-generation and later-change failure pointsthe skill has no such rule · Mode 2 — audit, check 4: MECE	present	present
Suggested first-level structure uses causes, symptoms, constraints, or topics instead of actionsthe skill has the rule; the model did not follow it · Core loop, §4 Groups	present	present
Inline annotation, capped findings, five-axis score, or three highest-value fixes are absentthe skill has the rule; the model did not follow it · Mode 2 — audit, delivery contract	present	absent
Unsupported evidence requirements, process claims, or implementation specifics are introducedthe skill has the rule; the model did not follow it · Mode 2 finding rule and Core loop §8	present	absent

Decisive questions

Did the audit state precisely that the required changes are absent?
without no—it calls the recommendation buried rather than absent
with Minto partial
Did it distinguish a missing answer from a merely buried answer?
without no
with Minto no
Did it identify streamlining as an undeveloped conclusion?
without no
with Minto no—it treats streamlining as the recommendation
Did it name incomplete process logic and unseparated failure points?
without no
with Minto no
If a replacement structure was suggested, is its key line made of actions?
without no—the suggested categories are problem topics
with Minto no
What false positives or invented details appear?
without invents immediacy, recurrence, a task force, and a 30-day deadline
with Minto proposes cause and capacity branches rather than corrective actions
Did the output remain diagnosis-only and include the required audit apparatus?
without diagnosis-only, but the required apparatus is absent
with Minto yes

Judge’s check	without	with Minto
First-level kind matches the gold	yes	yes
Invented facts	yes	no
Source lost without a note	no	no
Mode respected	yes	yes
Order type named	yes	no

4 more judge checks do not apply to mode audit

How the test was built

qwen3.6-fp8 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 2 424 B

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

<!-- Faithful paraphrase of Minto (2010), Appendix B, Exhibits B-5 and B-6.
     Original wording for this fixture. -->

# Fixture 03: Period Graph Books

**Mode:** `audit`
**Language:** `en`

## Context

Corporate Financial Analysis recently took over production of four monthly
graph books used in presentations to senior management. Management knows that
errors and inconsistencies have appeared and asked for corrective proposals.

## Before

```text
Subject: Period Graph Books

Since Period 5, Financial Analysis has been responsible for producing four
graph-presentation books previously handled by Corporate Planning. This memo
describes issues that appeared after the transfer.

Production

1. Gather data from divisional forms, internal documents, and telephone calls.
2. Calculate graph points manually or on the corporate computer, including
   rolling values for revenue, cost, and percentages.
3. Transcribe the points to input sheets. Analysts update the current period,
   then Color Graphics re-enters the points into its own database and generates
   the graphs.
4. Check the data and calculations.

Issues

The transfer inserted another person into an already fragmented process.
Divisional staff and three corporate departments participate, and the same
information may be entered into computer systems three times.

For one book, divisional data are loaded into a dedicated corporate program,
calculated, transferred to Color Graphics input sheets, entered again, and
returned for updates. Problems include inconsistent data between periods and
regions, incorrect variable-cost calculations, unexplained changes to points
that were previously correct, and prior-period data that must be re-entered.

Overall assessment

Most problems arise from a cumbersome fragmented process. No one controls the
data from beginning to end, responsibility is unclear, and errors can fall
between stages. The production process badly needs streamlining, both for each
book and for common calculations. The current team cannot manage the required
changes as well as ongoing production.

Could we discuss how best to proceed?
```

## Task

Audit this document against the Minto Pyramid Principle. Diagnose the defects
and explain what must change. Do not rewrite the document.

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

Prompt with Minto five parts, in order · 29 997 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 03: Period Graph Books

Mode: audit
Language: en

Context

Corporate Financial Analysis recently took over production of four monthly
graph books used in presentations to senior management. Management knows that
errors and inconsistencies have appeared and asked for corrective proposals.

Before

Subject: Period Graph Books
                
                Since Period 5, Financial Analysis has been responsible for producing four
                graph-presentation books previously handled by Corporate Planning. This memo
                describes issues that appeared after the transfer.
                
                Production
                
                1. Gather data from divisional forms, internal documents, and telephone calls.
                2. Calculate graph points manually or on the corporate computer, including
                   rolling values for revenue, cost, and percentages.
                3. Transcribe the points to input sheets. Analysts update the current period,
                   then Color Graphics re-enters the points into its own database and generates
                   the graphs.
                4. Check the data and calculations.
                
                Issues
                
                The transfer inserted another person into an already fragmented process.
                Divisional staff and three corporate departments participate, and the same
                information may be entered into computer systems three times.
                
                For one book, divisional data are loaded into a dedicated corporate program,
                calculated, transferred to Color Graphics input sheets, entered again, and
                returned for updates. Problems include inconsistent data between periods and
                regions, incorrect variable-cost calculations, unexplained changes to points
                that were previously correct, and prior-period data that must be re-entered.
                
                Overall assessment
                
                Most problems arise from a cumbersome fragmented process. No one controls the
                data from beginning to end, responsibility is unclear, and errors can fall
                between stages. The production process badly needs streamlining, both for each
                book and for common calculations. The current team cannot manage the required
                changes as well as ongoing production.
                
                Could we discuss how best to proceed?

Task

Audit this document against the Minto Pyramid Principle. Diagnose the defects
and explain what must change. Do not rewrite the document.

The judge’s gold gold.md

Gold 03: Period Graph Books

Source structure: Minto (2010), Appendix B, Exhibits B-5 and B-6.

Book-derived defects

Defect	Severity
The document never states the changes required; it therefore does not answer management's question	Critical
It describes the current process instead of presenting a top and key line	Critical
"The purpose is to describe issues" is a topic, not an answer	Critical
The process description is incomplete and does not isolate each system's failure points	Medium
Problems in graph-point generation and later presenter changes are not clearly separated	Medium

The buried conclusion is that the process needs streamlining, but this is not
turned into an actionable answer.

Expected replacement structure

Reader question: What changes should we make to prevent unreliable graphs?

Top: Remove the opportunities for error across data transfer, graph-point
generation, and final presentation changes.

Key-line kind: changes/actions.

Create a direct data link to the corporate computer.
Create a reliable computerized routine for graph-point generation.
Require presenter changes to be revalidated before use.

Audit outputs should name the missing answer, distinguish it from a conclusion
that is merely buried, annotate only visible evidence, provide no rewritten
memo, and avoid inventing defects in unseen material.

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