planning-stories

Decomposes PRDs and system architecture into user story outlines with dependencies and shared contracts.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/gleeb/ai-registry --skill planning-stories
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: planning-stories
Source: https://github.com/gleeb/ai-registry/tree/main/systems/sdlc/skills/planning-stories
Command: npx skills add https://github.com/gleeb/ai-registry --skill planning-stories

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill turns a validated PRD and system architecture into a structured set of user story outlines, so execution agents can implement features without ambiguous scope, missing dependencies, or duplicated contract definitions.

Core Features & Use Cases

  • Story Decomposition with Guardrails: Breaks each PRD user story group into right-sized stories (bounded for ~30–60 minutes) while enforcing strict “WHEN / WHEN NOT to use” rules to prevent premature or out-of-scope work.
  • Dependency Manifest + Integration Strategy: Produces per-story dependency manifests and an integration strategy that records mocks vs real connections and ensures every mock is realized later or explicitly deferred.
  • Shared Contract Identification: Detects shared interfaces (data shapes, API contracts, auth models, error formats), extracts them into plan/contracts/, and connects stories to provided/consumed contracts.
  • Folder Structure Output: Creates plan/user-stories/US-NNN-name/ folders with story.md files and adds plan/contracts/ entries for identified shared contracts.

Quick Start

Use planning-stories to decompose plan/prd.md and plan/system-architecture.md into plan/user-stories/ story folders after the Planning Hub validates the PRD and architecture inputs.

Frequently Asked Questions about planning-stories

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

FAQPage Schema
How do I decompose a PRD into implementable user stories?

To decompose a PRD into implementable user stories, you need a validated PRD and system architecture to generate bounded story outlines with dependency manifests, shared contract identification, and folder structure outputs.

What is shared contract extraction in user story planning?

Shared contract extraction in user story planning identifies shared interfaces like API contracts and data shapes, extracts them into a shared registry format under `plan/contracts/`, and connects stories to provided or consumed contracts.

How do I prevent circular dependencies when breaking down system architecture into stories?

To prevent circular dependencies when breaking down system architecture into stories, apply ordering validation during story decomposition to detect boundary misalignment and ensure per-story dependency manifests enforce correct execution sequencing.

Do I need a validated PRD before generating user story outlines?

Yes, you need a validated PRD before generating user story outlines because strict PRD and architecture validation gating is required to prevent premature or out-of-scope work during Phase 2 story decomposition.

How do mock and realize integration strategies work in story decomposition?

Mock and realize integration strategies in story decomposition record mock versus real connections per story, ensuring every mock is either realized later or explicitly deferred to maintain integration consistency across user stories.

What is the best way to structure user stories for execution agents?

The best way to structure user stories for execution agents is creating `plan/user-stories/US-NNN-name/` folders with `story.md` files, bounding stories to 30–60 minutes of work with explicit dependency manifests to eliminate ambiguous scope.