Agent Creating

Create agent manifest files in .claude/agents with YAML headers and workflows.

Updated Nov 13, 2025
One-click install
npx skills add https://github.com/gzbomerif-sketch/vibed-Web-Start --skill agent-creating-gzbomerif-sketch
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Agent Creating
Source: https://github.com/gzbomerif-sketch/vibed-Web-Start/tree/main/.claude/skills/agent-builder
Command: npx skills add https://github.com/gzbomerif-sketch/vibed-Web-Start --skill agent-creating-gzbomerif-sketch

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires context7, mcp-api, python>=3.8.

What problem does it solve?

This Skill streamlines the creation of reusable Claude agents, saving time and ensuring consistent quality across agents.

Core Features & Use Cases

  • Agent scaffolding: Automatically creates a new agent file under .claude/agents with a name-derived filename and a structured description.
  • Quality guardrails: Ensures the agent description is concise (2-4 sentences) and defines a clear persona and goal.
  • Workflow guidance: Provides a minimal, actionable activation workflow for using the new agent.
  • File organization: Advises on persona, goals, prompts, and samples for quick reuse.

Quick Start

When asked to create a new agent, I'll generate a file in .claude/agents named after the agent (e.g., "code-reviewer.md" or "stripe-implementor.md"), populate it with a clear description, persona, and goal, and supply a compact activation workflow. The agent will be ready to load into your agent runtime with minimal edits.

Frequently Asked Questions about Agent Creating

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

FAQPage Schema
How do I create a reusable AI agent to automate workflow tasks?

Create a reusable AI agent by defining a clear persona, goal, and 2–4 sentence description, then use this Skill to generate a structured Markdown manifest in .claude/agents with a YAML header (name, description, model, tools) and actionable workflow instructions for immediate deployment.

What should an agent description include to be production-ready?

An agent description requires a defined persona, explicit goal statement, concise 2–4 sentence overview, minimal actionable Markdown workflow steps, and reference to convexGuidelines.md. The Skill scaffolds these elements into a well-structured manifest file ready to load into your agent runtime.

Can I use Claude agents to automate code reviews or API integrations?

Yes. You can create specialized Claude agents for tasks like code-reviewer or stripe-implementor by specifying the agent name and role. This Skill generates the agent manifest with persona, goal, and workflow guidance tailored to your automation use case.

What file structure does the agent creation process follow?

Agents are created as Markdown files in .claude/agents with names derived from the agent purpose (e.g., code-reviewer.md). Each file contains a YAML header with metadata and a body detailing persona, goal, and activation workflow for consistent organization and quick reuse.

Do I need to manually configure tools and model selection for new agents?

The Skill generates agents with tool and model fields in the YAML header, ensuring consistent structure. You define the agent's purpose and specifications upfront; the Skill scaffolds the manifest to minimize manual setup and enable immediate usability.

How does agent creation ensure quality and consistency across multiple agents?

The Skill enforces quality guardrails by requiring concise descriptions (2–4 sentences), explicit persona and goal definition, and structured Markdown templates. This standardized approach maintains consistency across agents and reduces creation time for reusable automation workflows.