What makes these documents 'AI artifacts'?

These documents use specific frontmatter, hierarchical linking, and length constraints (max 800 lines) specifically designed to be loaded into an AI agent's context without causing performance degradation or context loss.

Does this skill support specific software architectures?

Yes, it is optimized for layered architecture patterns (MP) including distinct layers for routers, services, repositories, and database schemas.

Why does the skill split files at 800 lines?

Large files consume significant tokens and can lead to 'middle-of-the-document' forgetfulness in LLMs. Splitting files ensures the agent focuses on high-density, relevant information for the current phase.

How does subagent assignment work?

The skill analyzes the nature of each task—such as UI components or database migrations—and automatically maps them to specialized roles like 'ui-engineer-enhanced' or 'data-layer-expert' in the implementation plan.

Planning & Project Architecture

Name: Planning & Project Architecture
Author: miethe

bymiethe

Productivity & Workflow

Generates and optimizes PRDs, phased implementation plans, and progress tracking artifacts optimized for AI agent context.

About

The Planning Skill is a specialized tool for Claude Code designed to translate high-level feature requests into structured, AI-readable technical documentation. It automates the creation of Product Requirements Documents (PRDs) and detailed implementation plans while strictly adhering to token-efficiency standards by splitting documents longer than 800 lines. By organizing tasks into logical phases and assigning specialized subagents—such as data-layer experts or UI engineers—it ensures that complex software projects remain manageable and context-rich for both human developers and AI coding assistants.

Key Features

Integrated progress tracking following CLAUDE.md policy
Automated PRD generation using standardized architectural templates
Token-efficient document optimization via modular file splitting
Support for layered architecture patterns (Routers, Services, Repositories)
Phased implementation planning with specific subagent task assignments
0 GitHub stars

Use Cases

Setting up structured task tracking for multi-agent development workflows
Refactoring massive planning documents into smaller, token-efficient modules
Converting a brief feature description into a multi-phase technical roadmap

About

Key Features

Integrated progress tracking following CLAUDE.md policy
Automated PRD generation using standardized architectural templates
Token-efficient document optimization via modular file splitting
Support for layered architecture patterns (Routers, Services, Repositories)
Phased implementation planning with specific subagent task assignments
0 GitHub stars

Use Cases

Setting up structured task tracking for multi-agent development workflows
Refactoring massive planning documents into smaller, token-efficient modules
Converting a brief feature description into a multi-phase technical roadmap