Codex gpt-5.6-terra на кейсе «Набор ежемесячных отчётов»: с Minto без изменений

Структура 5 → 5 из 8 0 Качество 6 → 6 из 10 0

Судья об ответе с Minto

It offers the strongest process-level annotation, though it never isolates graph-point generation from later changes.

Режим audit · кейс 03-period-graph-books · прогон v1.7.0-final

Структура из 8, по 2 балла на ось
СтруктураБез Minto 5С Minto 5Изменение
Вершина21−1
Состав ключевой линии11
Уровни12+1
Порядок и род11
Качество из 10, по 2 балла на ось
КачествоБез Minto 6С Minto 6Изменение
Вершина21−1
Однородные группы11
Объяснимый порядок11
MECE11
Видимая соразмерная структура12+1

Ответы моделей и слова судьи приведены в оригинале, по-английски.

Без Minto

контрольный промпт · слепая метка out-08

Жёстких провалов нет

It diagnoses the absent recommendation cleanly, but its replacement structure is still reason-led and includes unsupported design choices.

Minto audit — Period Graph Books

Overall diagnosis

The document does not yet form a persuasive pyramid. It describes the production process and its problems, but it never states a clear governing answer or corrective recommendation. The reader must infer both the core problem and the required decision.

Defects
  • No answer at the top. “Could we discuss how best to proceed?” is a request for discussion, not a recommendation or decision statement.
  • Situation and complication are mixed. The transfer of responsibility, process description, examples of re-entry, and error types are interwoven rather than leading cleanly to one defined complication.
  • The key question is unstated. The memo implies: “What changes should management authorize to make graph-book production reliable and controllable?” It should be explicit.
  • The main conclusion is too broad. “Streamlining” and “required changes” do not specify what should change, who should own it, or what management must approve.
  • Reasons are not structured as a logical governing set. Fragmentation, duplicate entry, calculation defects, data inconsistency, and unclear responsibility overlap and appear as a list of observations.
  • Evidence is mixed with assertions. Specific defects—such as incorrect variable-cost calculations and unexplained point changes—should support defined arguments, not sit beside them as parallel main points.
  • The production sequence is over-detailed for the opening. It is useful diagnostic evidence but does not establish a management-level message on its own.
  • No decision path is provided. The document says the current team cannot both maintain production and make changes, but does not identify the needed resources, ownership model, or implementation mandate.
What must change
  • State one decisive recommendation first: the production process requires a defined redesign and a single accountable owner, with capacity separate from ongoing monthly production.
  • Frame the opening as a clear situation–complication–question–answer sequence.
  • Organize supporting points into a small, non-overlapping set of reasons, such as:
    • the process creates avoidable data duplication and handoffs;
    • controls and calculation standards are insufficient;
    • end-to-end accountability and implementation capacity are missing.
  • Place process steps, system re-entry examples, and individual calculation errors beneath the relevant supporting reason as evidence.
  • Specify the management action sought: authorize the redesign, assign accountable ownership, and provide dedicated implementation resources.
  • Distinguish immediate controls needed to protect current production from the longer-term process redesign.

С Minto

промпт с навыком · слепая метка out-02

Жёстких провалов нет

It offers the strongest process-level annotation, though it never isolates graph-point generation from later changes.

Legend: ▲ top / answer · ● first-level group · ○ support · ⚠ structural violation · ↑ promote · ⇄ wrong order · ⊗ overlap · ⊕ gap · ✂ cut or move to appendix

Subject: Period Graph Books [▲ topic, not an answer]
                
                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. [⚠ purpose statement; it does
                not state the reader’s decision or the answer]
                
                Production [⚠ process description begins before the document’s conclusion]
                
                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. [⊕ incomplete control loop: no clear owner,
                   hand-off approval, or correction/reconciliation stage]
                
                Issues [⚠ mixed group: process causes, organizational observations, and error
                symptoms]
                
                The transfer inserted another person into an already fragmented process. [● cause]
                Divisional staff and three corporate departments participate, and the same
                information may be entered into computer systems three times. [○ supporting detail]
                
                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. [○ supporting detail]
                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. [⚠ mixed kinds; these
                symptoms need to support distinct conclusions]
                
                Overall assessment [↑ the document’s answer is buried here]
                
                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. [⊗ overlapping causal claims: fragmented process, absent
                end-to-end control, and unclear responsibility partly describe the same defect]
                The production process badly needs streamlining, both for each book and for
                common calculations. [↑ actionable conclusion belongs at the top; ⊕ “streamlining”
                does not specify the changes requested]
                The current team cannot manage the required changes as well as ongoing production.
                [⊕ no proposal for ownership, capacity, or decision]
                
                Could we discuss how best to proceed? [⚠ vague ask; management needs a defined
                decision or proposal]
LocationViolationFix
Subject and openingNo answer to management’s implicit question: what corrective action should be approved?Open with one recommendation, not a topic or statement of purpose.
“Overall assessment”The core conclusion and recommendation are buried after process detail.Promote the conclusion into the first line; place facts beneath the changes they justify.
“Issues”Mixed kinds: causes, process mechanics, and error symptoms sit as one group.Separate same-kind branches, such as control/ownership failures, duplicate handling, and calculation/data-integrity failures.
Overall causal diagnosisOverlap between fragmented process, no end-to-end control, and unclear responsibility.Define non-overlapping problem branches and assign each evidence once.
Production stepsThe process is visible, but it does not support a complete corrective case.Use it as supporting evidence under each diagnosed failure; identify the missing control or ownership at each hand-off.
Recommendation“Needs streamlining” is too vague to close a decision.State the required actions, including ownership, single-source data flow, standard calculations, and implementation capacity—where supported; otherwise mark gaps.
Closing questionThe requested management action is undefined.End with a specific approval, resource, or direction required.

Score: 4/10 — baseline workable

AxisScoreAssessment
Top0/2No answer appears at the start.
Same-kind groups1/2Production is a time sequence; “Issues” is mixed.
Order1/2Production is chronological, but the overall argument is not.
MECE1/2Causal claims overlap; corrective categories are missing.
Display1/2Headings help, but the argument skeleton is not visible.

The three highest-value fixes:

  1. State a single management recommendation first, answering what should be approved now.
  2. Rebuild the body into 3–4 non-overlapping corrective changes, with process facts and error examples as support.
  3. Replace the vague discussion request with a defined decision on ownership, redesign scope, and implementation capacity.

Как устроен тест

Codex gpt-5.6-terra получила одну и ту же задачу дважды. Контрольный промпт — одна вводная строка и задача. Промпт с Minto — та же задача, перед которой стоит полный текст навыка. Судья читал оба ответа среди 16 ответов под слепыми метками и не знал, в каком из них был навык.

Промпт без Minto 2 424 Б
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.
Промпт с Minto пять частей по порядку · 29 997 Б
  1. Below is a skill written as an instruction. Read it in full and apply it to the task at the end.
  2. ===== SKILL.md ===== SKILL.md на f4813ce 17 235 Б
  3. ===== references/rules.md ===== references/rules.md на f4813ce 6 464 Б
  4. ===== references/templates.md ===== references/templates.md на f4813ce 3 742 Б
  5. ===== TASK ===== вход кейса целиком
  6. Return only the result, with no explanation of how you produced it.
Вход кейса before.md

<!-- 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
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.

Эталон судьи gold.md

Gold 03: Period Graph Books

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

Book-derived defects
DefectSeverity
The document never states the changes required; it therefore does not answer management's questionCritical
It describes the current process instead of presenting a top and key lineCritical
"The purpose is to describe issues" is a topic, not an answerCritical
The process description is incomplete and does not isolate each system's failure pointsMedium
Problems in graph-point generation and later presenter changes are not clearly separatedMedium

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.

  1. Create a direct data link to the corporate computer.
  2. Create a reliable computerized routine for graph-point generation.
  3. 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.

Судья gpt-5.6-sol, effort high · Codex gpt-5.6-terra, effort low · навык с коммита f4813ce · прогон v1.7.0-final