prompt-architect

Design, draft, test, and iterate on Claude Code skills, slash commands, and subagents.

Updated Jun 3, 2026
One-click install
npx skills add https://github.com/SunshinePeace213/soriza-startup-layer --skill prompt-architect-sunshinepeace213
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prompt-architect
Source: https://github.com/SunshinePeace213/soriza-startup-layer/tree/main/.claude/skills/prompt-architect
Command: npx skills add https://github.com/SunshinePeace213/soriza-startup-layer --skill prompt-architect-sunshinepeace213

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires claude-code, python, yaml, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides a comprehensive solution for designing, drafting, testing, and iterating on Claude Code skills, slash commands, and subagents, streamlining the full artifact lifecycle.

Core Features & Use Cases

  • End-to-End Development: Supports the full lifecycle of building Claude Code artifacts, including design, draft, review, place, validate, trigger, and ship.
  • Artifact Types: Covers skills (directory format), legacy single-file commands, and subagents.
  • Use Case: Use this Skill to create a new skill for code review, draft a SQL migration skill, or build a subagent for security reviews.

Quick Start

To design a new skill for code review, use the following command: /prompt-architect I want to create a skill for code review.

Frequently Asked Questions about prompt-architect

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

FAQPage Schema
How do I build and test Claude Code skills end to end?

Building Claude Code skills involves a full artifact lifecycle: design, draft, review, place, validate, trigger, and ship. This process manages directory-format skills, subagents, and slash commands within the Claude Code environment.

What is the best way to create a subagent for code review?

Creating a subagent for code review involves drafting the agent logic and iterating through validation triggers. The full artifact lifecycle supports building subagents specifically tailored for tasks like security reviews or automated checks.

Does the skill development lifecycle support legacy single-file commands?

Yes, the skill development lifecycle supports legacy single-file commands alongside directory-format skills and subagents. You can draft, validate, and ship these legacy slash commands using the same end-to-end workflow.

Do I need Python and YAML to draft Claude Code subagents?

Yes, Python and YAML are required dependencies to draft and iterate on Claude Code subagents and skills. These technologies drive the scripts and references required for the artifact validation and placement process.

Why does my slash command fail validation during the artifact lifecycle?

Slash command validation failures during the artifact lifecycle often occur when scripts or YAML configurations are incorrectly placed. You must review the artifact, fix structural issues, and re-trigger the validation step.