skills.rest
hb9397
hb9397COMMUNITY
Explore all skills by this author →
GitHub

custom-skill-design

Design, develop, and validate AI Agent Skills for Codex and Claude runtimes.

Updated Mar 5, 2026
Search Tags:#codex#claude#skill-development#ai-agent#workflow-automation#prompt-engineering
One-click install
npx skills add https://github.com/hb9397/ai-agent-harness-docs --skill custom-skill-design
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: custom-skill-design
Source: https://github.com/hb9397/ai-agent-harness-docs/tree/main/maintainer/skills/custom-skill-design
Command: npx skills add https://github.com/hb9397/ai-agent-harness-docs --skill custom-skill-design

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill streamlines the complex process of creating, testing, and refining AI Agent Skills, ensuring they are robust, platform-agnostic, and trigger-accurate for both Codex and Claude environments.

Core Features & Use Cases

  • Full Lifecycle Management: Guides you from initial interview and design to implementation, testing, and description optimization.
  • Platform-Agnostic Design: Ensures Skills work seamlessly across different AI runtimes while respecting platform-specific metadata requirements.
  • Use Case: When you have a recurring manual workflow and want to turn it into a reliable, reusable AI Skill, this tool manages the file structure, testing loops, and trigger logic to ensure it works perfectly every time.

Quick Start

Use the custom-skill-design skill to start a new project for designing an automated code review agent.

Frequently Asked Questions about custom-skill-design

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

FAQPage Schema
How do I build an AI Agent Skill that works across both Claude and Codex runtimes?▼

Platform-agnostic AI Agent Skill development requires a structured lifecycle framework to design modular components like prompts, templates, and deterministic scripts. This approach ensures robust cross-runtime execution while respecting platform-specific metadata for both Claude and Codex environments.

What is the best way to structure prompt engineering and testing loops for AI workflows?▼

Structured prompt engineering relies on automated evaluation loops within an iterative testing framework to validate triggers and refine prompt templates. This ensures robust execution and accurate workflow automation through continuous component testing.

How do I turn a recurring manual workflow into a reusable automated AI agent?▼

Converting a manual workflow into a reusable AI agent involves a structured interview process to design triggers, implement modular components, and run iterative testing loops. This lifecycle management ensures the final automated agent executes reliably across platforms.

Do I need specific dependencies or environments to design modular skill components?▼

No external dependencies are required to design modular skill components. The framework provides built-in support for generating file structures containing scripts, references, and assets needed to validate deterministic execution for your AI agents.

Why are my AI agent skill triggers failing to execute accurately during workflow automation?▼

Inaccurate skill triggers often result from skipping automated evaluation loops and strict design principles during development. Implementing an iterative testing framework validates trigger logic and refines prompt templates to ensure robust execution across runtime environments.

Related Skills

Explore More Software Engineering Skills→
s977043s977043
2

skill-creator

Design a Claude Code skill from a use case and produce a repo-ready package.

Community
Advanced
hainamchunghainamchung
54

agent-assistant-deploy

Route deployment tasks to the Agent Assistant workflow via /deploy in Codex-based agents.

Community
Intermediate
zhangzhang-111-izhangzhang-111-i
2

prompt-engineer-toolkit

Evaluate and version AI prompts with A/B testing and diffing.

Community
Advanced
devCharuzudevCharuzu

senior-prompt-engineer

Design prompts and evaluation frameworks for LLM workflows.

Community
Advanced