skill-decomposer

Extract a single reusable capability from a larger skill or workflow.

20|3|Updated May 14, 2026
One-click install
npx skills add https://github.com/cyberAlchemyAI/Arcanum --skill skill-decomposer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-decomposer
Source: https://github.com/cyberAlchemyAI/Arcanum/tree/main/.claude/skills/skill-decomposer
Command: npx skills add https://github.com/cyberAlchemyAI/Arcanum --skill skill-decomposer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Skill Decomposer is a tool to extract a single reusable capability from a larger source, avoiding over-conversion and enabling focused sigil development.

Core Features & Use Cases

  • Map Source: identify source scope, context, and gates.
  • Extract: define one target capability and its reusable core.
  • Boundary: decide what belongs inside or outside the candidate sigil.
  • Handoff: produce a package plan for skill-transcriptor.
  • Validate: verify the candidate is coherent, reusable, and not source-coupled.
  • Observe: emit decomposition telemetry for reflection.

Quick Start

Provide a source fragment and allow Skill Decomposer to identify candidate sigils and prepare a handoff package when boundaries are clear.

Frequently Asked Questions about skill-decomposer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract a reusable capability from a tangled workflow without importing local machinery?

Skill decomposition isolates a single reusable capability from a larger source by mapping scope, identifying candidates, and defining boundaries. This prevents over-conversion and ensures the extracted capability remains decoupled from local machinery.

What is the best way to isolate reusable sigils from a prompt containing multiple potential capabilities?

Isolating reusable sigils from a source with multiple capabilities requires mapping candidate discovery points and making boundary decisions. You validate the extracted candidate to ensure it is coherent and not source-coupled before generating a handoff package.

How do I prepare a handoff package for skill-transcriptor after defining capability boundaries?

Preparing a handoff package for skill-transcriptor requires executing map, extract, boundary, and handoff steps on the source. The process outputs a structured decomposition report and a transferable package plan for the downstream transcriptor.

Do I need a SKILL.md frontmatter at the root to perform capability extraction and boundary mapping?

Yes, a SKILL.md frontmatter at the root is required to execute the map, extract, boundary, and handoff steps. This frontmatter provides the necessary structural context to initiate the decomposition process and generate the handoff package.

When should I not use a decomposition approach to extract capabilities from a larger skill?

You should avoid a decomposition approach when the source contains a single clear capability or when boundary decisions are unambiguous. Decomposition is designed specifically for tangled sources with multiple potential sigils and unclear boundary demarcations.