How do I install the Problem Decomposition skill?

Run npx skillmds@latest add npbuilds/problem-decomposition in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Problem Decomposition safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Problem Decomposition?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Problem Decomposition free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Problem Decomposition?

npbuilds (@npbuilds) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Problem Decomposition by npbuilds — Agent Skill

Name: Problem Decomposition
Author: npbuilds

Problem Decomposition

Route a diagnosed problem to the right decomposition method — root-cause technique selection and the Polya heuristic catalog. Activate at six-eyes Phase 3 (Decompose), after a typology profile exists, when you need to break a statement into atomic, root-tagged parts before reframing or auditing. This director owns the "break it down to its real components" question.

npbuilds 01c6f73 2.4 KB Updated Sep 17, 2026

File contents

Problem Decomposition — Director

A diagnosed problem is rarely atomic. This director realizes six-eyes' Phase 3 (Decompose): it selects the right root-cause method for the problem's typology and stakes, tags each piece as root vs. symptom, and grounds the work in Polya's problem-solving heuristics. The goal is a set of atomic, root-tagged claims that downstream reframing and audit can operate on cleanly.

Decomposition partners with the research domain's claim-decomposer (cross-domain) for atomic-claim extraction; this director focuses on root-cause structure — which split is causal, which is cosmetic.

Routing Table

Child skill	Use when	Output
`root-vs-symptom-tagger`	You need to pick a root-cause method (5 Whys / Fishbone / dependency map / Current Reality Tree) and tag each part root vs. symptom	Method choice + per-claim root/symptom tags
`polya-method`	You need the heuristic basis for understand → plan → execute → look-back, or specialize/generalize moves	Reference: Polya's How to Solve It catalog

Routing Logic

Take the typology tag from problem-diagnosis. It selects the method: dependency mapping for mess, Current Reality Tree for wicked, 5 Whys / Fishbone for well- or ill-defined. root-vs-symptom-tagger performs this selection.
Pull polya-method when the decomposition stalls — its specialize/generalize operations and the "understand the problem" step often unstick a tangled statement.
If decomposition yields 15+ atomic claims, hand back to six-eyes to ask the user to prioritize before continuing.

Scope Boundaries

In scope: selecting a decomposition method, root vs. symptom tagging, Polya heuristics.
Out of scope: classifying the problem (→ problem-diagnosis), generating alternative formulations (→ problem-reframing).
Cross-domain: atomic-claim extraction proper is research/claim-decomposer; this director routes to it via the orchestrator rather than duplicating it.

---name: problem-decompositiondescription: Route a diagnosed problem to the right decomposition method — root-cause technique selection and the Polya heuristic catalog. Activate at six-eyes Phase 3 (Decompose), after a typology profile exists, when you need to break a statement into atomic, root-tagged parts before reframing or auditing. This director owns the "break it down to its real components" question.---# Problem Decomposition — DirectorA diagnosed problem is rarely atomic. This director realizes six-eyes' Phase 3 (Decompose): itselects the right root-cause method for the problem's typology and stakes, tags each piece asroot vs. symptom, and grounds the work in Polya's problem-solving heuristics. The goal is a setof atomic, root-tagged claims that downstream reframing and audit can operate on cleanly.Decomposition partners with the research domain's `claim-decomposer` (cross-domain) for atomic-claimextraction; this director focuses on *root-cause structure* — which split is causal, which iscosmetic.## Routing Table| Child skill | Use when | Output ||---|---|---|| `root-vs-symptom-tagger` | You need to pick a root-cause method (5 Whys / Fishbone / dependency map / Current Reality Tree) and tag each part root vs. symptom | Method choice + per-claim root/symptom tags || `polya-method` | You need the heuristic basis for understand → plan → execute → look-back, or specialize/generalize moves | Reference: Polya's *How to Solve It* catalog |## Routing Logic1. Take the typology tag from `problem-diagnosis`. It selects the method: dependency mapping for **mess**, Current Reality Tree for **wicked**, 5 Whys / Fishbone for **well-** or **ill-defined**. `root-vs-symptom-tagger` performs this selection.2. Pull `polya-method` when the decomposition stalls — its specialize/generalize operations and the "understand the problem" step often unstick a tangled statement.3. If decomposition yields 15+ atomic claims, hand back to `six-eyes` to ask the user to prioritize before continuing.## Scope Boundaries- **In scope:** selecting a decomposition method, root vs. symptom tagging, Polya heuristics.- **Out of scope:** classifying the problem (→ `problem-diagnosis`), generating alternative formulations (→ `problem-reframing`).- **Cross-domain:** atomic-claim extraction proper is `research/claim-decomposer`; this director routes to it via the orchestrator rather than duplicating it.

Frequently asked questions