decompose-intent

Decompose de-risked intents into child intents or shippable leaf specs.

15|5|Updated May 4, 2026
One-click install
npx skills add https://github.com/eugenelim/agent-ready-repo --skill decompose-intent
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: decompose-intent
Source: https://github.com/eugenelim/agent-ready-repo/tree/main/packs/product-engineering/.apm/skills/decompose-intent
Command: npx skills add https://github.com/eugenelim/agent-ready-repo --skill decompose-intent

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you break a validated intent into the next actionable level without skipping product or delivery rungs, so the tree stays honest and buildable.

Core Features & Use Cases

  • Recursively decomposes one level at a time into child intents or a shippable leaf spec.
  • Cuts leaf work by shippability, not by component or layer, so each slice can ship and test on its own.
  • Records the decomposition rationale and supports tracker projection for markdown, Linear, or Jira Align workflows.
  • Use it when a feature needs to be split into deliverable slices, when a larger bet needs child intents, or when a leaf should become a brief for a delivery loop.

Quick Start

Ask the skill to decompose the current de-risked intent into the next level down and capture the parent link, cut rationale, and any leaf projection needed for the target scale.

Frequently Asked Questions about decompose-intent

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

FAQPage Schema
How do I break down a validated product intent into buildable slices?

To break down a validated product intent into buildable slices, you recursively decompose the intent one level at a time into child intents or shippable leaf specs. This ensures the planning tree remains honest and actionable.

What is the best way to split features for tracker projection without skipping levels?

Splitting features for tracker projection without skipping levels requires cutting leaf work by shippability rather than by component. This approach allows each slice to ship and test independently while maintaining parent intent backlinks.

How does recursive intent decomposition work for shippable specs?

Recursive intent decomposition works by applying a step-by-step scale-aware projection to capability, feature, and leaf-level planning. It requires a surviving riskiest assumption and records the decomposition rationale to generate tracker-ready projections.

Can I use this intent decomposition approach with Jira Align and Linear workflows?

Yes, you can use this intent decomposition approach with Jira Align, Linear, or markdown workflows. It supports tracker projection by transforming leaf-level intents into briefs or per-component slices formatted for your specific target scale.

When do I need a surviving riskiest assumption for feature decomposition?

You need a surviving riskiest assumption for feature decomposition when moving from a de-risked intent to the next actionable level. This prerequisite ensures the decomposition rationale is backed by validated hypotheses before generating shippable specs.

Why decompose leaf work by shippability instead of by component layer?

Decomposing leaf work by shippability instead of by component layer ensures each slice can ship and test on its own. This prevents incomplete feature fragments and maintains an honest, buildable product planning tree.