agent-creator

Name: agent-creator
Availability: InStock
Author: fercracix33

Creates and refactors Claude Code agents with YAML frontmatter and skill invocations.

Updated Sep 29, 2025

Search Tags:#validation #claude-code #skill-scaffolding #yaml-frontmatter #agent-creation

Ready to run

npx skills add https://github.com/fercracix33/poli2-clean-architecture --skill agent-creator-fercracix33

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agent-creator
Source: https://github.com/fercracix33/poli2-clean-architecture/tree/main/.claude/skills/agent-creator
Command: npx skills add https://github.com/fercracix33/poli2-clean-architecture --skill agent-creator-fercracix33

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires bash, grep, sed, awk, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill eliminates the complexity and inconsistency of creating Claude Code agents by providing a structured 6-phase workflow that follows Anthropic's official best practices.

Core Features & Use Cases

Agent Creation: Guide users through designing minimal agent files (~150 lines) with proper YAML frontmatter and skill invocations.
Refactoring: Transform verbose, monolithic agent files into clean, minimal structures with comprehensive companion skills.
Use Case: When a user needs to create a new security-auditor agent, this Skill handles everything from discovery to validation and automatic handoff to skill-creator.

Quick Start

Use the agent-creator skill to create a new security-auditor agent for vulnerability scanning.

Frequently Asked Questions about agent-creator

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

FAQPage Schema

How do I create a Claude Code agent following Anthropic best practices?▼

Creating a Claude Code agent involves designing a minimal agent file (~150 lines) with proper YAML frontmatter, defining skill invocations, and following a 6-phase structured workflow. This Skill guides you through discovery, validation, and scaffolding to ensure compliance with Anthropic standards and automatic handoff to skill-creator for complete agent+skill packages.

What's included in a minimal agent structure for Claude Code?▼

A minimal agent structure requires approximately 150 lines with mandatory YAML frontmatter, skill invocations, proper metadata design, and validation checkpoints. This Skill automates scaffolding of agent directories, references, and the complete skeleton, ensuring your agent meets architectural requirements without manual boilerplate.

Can I refactor an existing agent file into a cleaner structure?▼

Yes. This Skill transforms verbose, monolithic agent files into clean, minimal structures by applying the 6-phase workflow and separating concerns into companion skills. The refactoring process maintains functionality while reducing complexity and establishing best-practice compliance.

How does the 6-phase agent development workflow work?▼

The workflow progresses through structured phases: discovery, design, validation, scaffolding, skill integration, and handoff. Each phase includes Context7 checkpoints to ensure correctness, with bash, grep, sed, and awk utilities automating validation and metadata generation throughout.

What dependencies and tools do I need to create agents with this Skill?▼

You need bash, grep, sed, and awk available in your environment. These command-line utilities power the automated validation, scaffolding, and metadata processing that enforce best practices and generate agent and skill directories during creation.

Do I need to know YAML frontmatter syntax to use this Skill?▼

No. This Skill automates YAML frontmatter design and validation as part of the scaffolding process, so you focus on agent logic and skill definition rather than syntax details. The Skill handles mandatory metadata fields and structure compliance.