How does this integrate with the rest of the AI-DLC repository?

It reads the application design state from aidlc-state.md and guides the user toward the construction phase once the units are successfully defined and approved.

Why does this skill require a separate planning phase?

The two-part stage ensures that the architectural strategy and unit boundaries are explicitly approved by the user before detailed artifacts are generated, preventing rework.

What specific artifacts are generated by this skill?

It generates unit definitions (unit-of-work.md), dependency matrices (unit-of-work-dependency.md), and story-to-unit maps (unit-of-work-story-map.md).

What is a 'unit of work' in this context?

A unit of work is a logical grouping of stories for development, representing either an independently deployable microservice or a specific module within a monolith.

AI-DLC System Unit Decomposition

Name: AI-DLC System Unit Decomposition
Author: yamatatsu

byyamatatsu

0•

Productivity & Workflow

Decomposes complex system designs into manageable units of work and dependency maps through a structured two-phase process.

This skill facilitates the transition from high-level application design to construction by breaking down systems into logical 'units of work' within the AI-DLC framework. It operates through a mandatory two-part workflow: a Planning phase to define unit boundaries, responsibilities, and code organization strategies, followed by a Generation phase that produces detailed documentation. By automating the creation of dependency matrices and story maps, it ensures architectural consistency and provides a clear roadmap for functional design and implementation in both microservice and monolithic environments.

Key Features

01Automated generation of dependency matrices and story maps

02Validation of architectural boundaries and unit responsibilities

03Structured two-part Planning and Generation workflow

040 GitHub stars

05Integrated audit logging and state management tracking

06Support for both greenfield projects and existing monoliths

Use Cases

01Mapping user stories to specific logical modules within a monolithic application

02Establishing a standardized code organization strategy for complex software projects

03Defining service boundaries and dependencies for a new microservices architecture

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yamatatsu/claude-aidlc inception-units-generation

For use in Claude.ai and ChatGPT

Download Skill