nw-agent-creation-workflow

Orchestrate a five-phase agent creation workflow from analysis to refinement.

Updated Apr 15, 2026
One-click install
npx skills add https://github.com/StudentCristian/nWave-github --skill nw-agent-creation-workflow
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: nw-agent-creation-workflow
Source: https://github.com/StudentCristian/nWave-github/tree/main/.github/skills/nw-agent-creation-workflow
Command: npx skills add https://github.com/StudentCristian/nWave-github --skill nw-agent-creation-workflow

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Detailed, repeatable guidance for building AI agents via a 5-phase process (ANALYZE -> DESIGN -> CREATE -> VALIDATE -> REFINE), enabling teams to start minimal and iteratively improve with clear gates.

Core Features & Use Cases

  • Phase-driven architecture with explicit inputs, outputs, and quality gates for each stage.
  • Deterministic workflow that reduces duplication and overlap with existing agents.
  • Clear governance around loading skills and validating artifacts before deployment.

Quick Start

Follow the five-phase workflow to analyze requirements, design architecture, create the agent and its skills, validate with the 14-point checklist, and refine based on feedback.

Frequently Asked Questions about nw-agent-creation-workflow

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

FAQPage Schema
What is a phase-driven agent creation workflow?

A phase-driven agent creation workflow orchestrates AI agent development through five sequential stages: ANALYZE, DESIGN, CREATE, VALIDATE, and REFINE. Each stage has explicit inputs, outputs, and quality gates to ensure safe, repeatable deployment and iterative improvement.

How do I validate AI agents before deployment?

You validate AI agents before deployment by applying a 14-point checklist during the VALIDATE phase of the workflow. This governance step checks frontmatter compliance, artifact organization, and skill-loading directives to ensure the agent meets all quality and safety requirements.

What's the best way to structure AI agent development to avoid duplication?

The best way to structure AI agent development and avoid duplication is using a deterministic, phase-sequenced workflow. It enforces architecture design and artifact organization rules that reduce overlap with existing agents before you create new ones.

Do I need specific dependencies to use a structured agent design process?

No specific dependencies are required to use this structured agent design process. The workflow is self-contained and relies on its own internal phase sequencing, frontmatter compliance rules, and validation gates rather than external libraries or components.

When should I not use a deterministic workflow for AI agent creation?

You should not use a deterministic workflow for AI agent creation when your project requires rapid, informal prototyping without strict governance. The explicit phase gates, frontmatter compliance, and validation checkpoints are designed for rigorous engineering, not ad-hoc builds.