agent-creator

Design and validate v2 architecture agents with automated quality checks.

4|1|Updated Jan 19, 2026
One-click install
npx skills add https://github.com/kimasplund/claude_cognitive_reasoning --skill agent-creator-kimasplund
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-creator
Source: https://github.com/kimasplund/claude_cognitive_reasoning/tree/main/skill-frameworks/agent-creator
Command: npx skills add https://github.com/kimasplund/claude_cognitive_reasoning --skill agent-creator-kimasplund

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This skill provides a structured, best-practice guide for creating domain-specific agents that adhere to the v2 architecture, helping teams design, validate, and deploy high-quality autonomous agents with repeatable patterns.

Core Features & Use Cases

  • Comprehensive learning resource for agent architecture, patterns, and quality rubric.
  • Step-by-step workflow covering requirements, phase design, validation, and deployment.
  • Reference integration with progressive disclosure to keep main files concise while depth is provided in references.

Quick Start

To scaffold a new agent, follow the 3-5 phase template in the guide, populate the frontmatter, implement the phase structure, and run the automated validator to ensure a quality score above the threshold before deployment.

Frequently Asked Questions about agent-creator

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

FAQPage Schema
How do I design production-ready autonomous agents?

Design production-ready autonomous agents by applying v2 architecture patterns, which structure requirement gathering, phase design, validation, and deployment to ensure measurable quality and robust tooling.

What is the v2 architecture pattern for agent creation?

The v2 architecture pattern for agent creation is a structured framework that uses phase templates, progressive disclosure, and automated validation to build high-quality domain-specific agents with repeatable patterns.

How do I validate autonomous agents before deployment?

Validate autonomous agents before deployment by running an automated validator to check phase structure, success criteria, and tool usage, ensuring the agent achieves a quality score above the deployment threshold.

Can I use progressive disclosure to keep agent files concise?

Yes, progressive disclosure keeps main agent files concise by separating core logic from deep reference materials, allowing the agent to access detailed information only when needed during specific phases.

What's the best way to scaffold a new domain-specific agent?

The best way to scaffold a domain-specific agent is to use a 3-5 phase template, populate the frontmatter, implement the phase structure, and run the automated validator to ensure quality before deployment.

Why does my agent design lack measurable quality criteria?

Agent design lacks measurable quality criteria when it does not implement v2 architecture patterns, which enforce success criteria, phase structure, and automated validation to guarantee robust deployment outcomes.