Skip to content
Skills Library
Management & product · Analysis

settle-question

/settle-question

Use to settle an open question whose deliverable is a judgment, with no runtime symptom to root-cause.

Repository
dude ↗
Licence
MIT
Last repository change
28.10.2026

Assessment and security

90/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content25 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
1
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 28.10.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Use to settle an open question whose deliverable is a judgment, with no runtime symptom to root-cause.

Original author description

Use to settle an open question whose deliverable is a judgment, with no runtime symptom to root-cause.

How it works

  1. Orchestration model
  2. Entry
  3. Explore
  4. Validate
  5. Synthesize
  6. Escalation
Original SKILL.md headings · 58 lines, 651 words
SKILL.md preview
---
name: settle-question
description: Use to settle an open question whose deliverable is a judgment, with no runtime symptom to root-cause.
---

# Settle Question

Settle an open question whose deliverable is a judgment. `superpowers:systematic-debugging` owns the core loop and runs unchanged here; this skill layers on only what a question-shaped run needs — splitting the question into points that can be settled, fixing what was measured, and reporting. Its Phase 4 Implementation does not run: an Investigation ends at findings.

## Orchestration model

- The points to settle are independent of each other, so fan out one worker per point. Each verdict stays with the orchestrator.
- A worker that measures, reads, or runs returns the command it ran and its raw output — or, for a read that ran no command, the file or record read and how its content was captured.
- **The worker that sweeps for absence or exhaustiveness is marked**: dispatch it at the `highest` tier in `using-dude`'s **Worker tier**. A sweep that stopped early hands back "nothing found", and neither the orchestrator nor any later reader of the report can tell that from an absence.

## Entry

A question, or an issue carrying the points to be settled. Closing any gap against `using-dude`'s **Understand** is Step 1's job — the first work here, not a precondition for starting.

## Explore

### Step 1: Split the question into settleable points

- Break the question into points that can each be settled on their own, and attach to each one what would settle it: the file, the record, the measurement, or the run whose result decides it.
- A point nothing can settle is not a point. It is a question to hand back to a person.

### Step 2: Fix the measurement base

- Record what was actually read: the revision, tree, or record. An installed tree and the repository it came from are different artifacts, and so are a branch and its base.
- Every claim of absence, exhaustiveness, or "only" names the sweep that produced it — run to completion and read whole, never a truncated or paged view of the output.

## Validate

### Step 3: Settle each point

- Run the core loop per `superpowers:systematic-debugging`: the points from Step 1 are the hypothesis pool, and what Step 1 attached to each point is the evidence that point owes.
- Separate evidence that settles a point from evidence that only suggests one answer. An unsettled point stays recorded as unsettled and is never promoted, per `using-dude`'s **Investigation workflow**.

## Synthesize

### Exit criteria

- Every point has a verdict with the decisive evidence behind it; or
- the point is recorded as unsettled, with what would settle it and who decides.

Once the report below is drafted, run `review-findings` on it for one pass, handing it the report; publish per `using-dude`'s **Stage boundaries**.

### Report format

1. **Verdict** — the answer to the question, in one paragraph.
2. **Measurement base** — the revision, tree, or record measured, and the sweep behind each claim of absence.
3. **Settled points** — per point, the verdict and the evidence that decided it.
4. **Alternatives rejected** — each one considered, with why it is wrong.
5. **Next items** — concrete next actions in order, each naming its owner (who decides) and its action type: file, decide, or measure. Not work to start here.
6. **Not measured** — the claims left unaudited, and what would settle each.

**Verdict** and **Next items** are what `using-dude`'s **Investigation → Change transition** carries forward — the next flow's input, never work this one starts. A question-shaped run has no reproduction to carry forward.

## Escalation

This run's stopping conditions are `using-dude`'s **Loop convergence**. One round is one core-loop pass over a point, and a finding repeats when a later pass faults the same point for the same reason. A finding that invalidates the question as asked is never settled in place: hand it back to the person who asked, per **Synthesize**'s second exit criterion.

Versions

The database has no other skills named settle-question.

Database snapshot: 13.11.2026 · package version e1d102672bcb (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 90decision-interview
    Settle each open branch of a plan or decision, one question per turn with a recommended answer; record mode keeps only pinned terms and qualifying decisions. USE WHEN named. NOT FOR unnamed clarifying, design, plans or building.
  • 90grill-and-records
    Settle a fuzzy plan, decision, or idea and write down what it settled: run the discussion, then write the terms into the glossary and the decisions into ADRs. It assembles; the discussion is `grilling`'s and each record is `adr`'s or `glossary`'s. Use when a decision needs a…
  • 100pm-scope
    Step 02 of the AI Product Management workflow. Use when someone says "scope this", "interview me on this", "pressure-test the plan", "what do we need to decide", "answer the open questions", "we need a decision on X", or when a build batch hits an undecided question. Resolves…
  • 100swipe-decide
    Use when the user wants to settle many open decisions quickly, says "let's decide these quickly" or "sort out the open decisions," or when 5+ open decisions have piled up in the conversation, or when a long unattended run (for example overnight) ends with questions deferred to…
  • 90write-diagrams
    Use when a document produced in this collection (board, plan, report, analysis, durable doc) could carry a diagram, or when a design question would settle faster shown than said. Picks the diagram that answers the open question and holds the discipline that keeps a diagram…
  • 90root-cause-priority
    用户面对多个相互关联的问题, 不知道先解决哪个时; 当需要区分根本原因 与表面症状时; 当问题有先后因果链需要追溯源头时。 不适用于: 只有一个简单问题的场景、问题之间没有因果关系的场景。

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.