create-agent

Generates custom agent files, slash commands, and memory folders for .NETworkspaces.

520|175|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/EvolutionAPI/evo-nexus --skill create-agent-evolutionapi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-agent
Source: https://github.com/EvolutionAPI/evo-nexus/tree/main/.claude/skills/create-agent
Command: npx skills add https://github.com/EvolutionAPI/evo-nexus --skill create-agent-evolutionapi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill guides users through creating a new custom agent for the workspace so they can extend automation, delegate domain-specific tasks, and standardize agent behavior without manual file layout or naming guesswork.

Core Features & Use Cases

  • Guided agent definition: Walks the user through domain, name, responsibilities, personality, model choice, color, and working folder.
  • File generation checklist: Specifies the required .claude/agents/ system prompt file, .claude/commands/ slash command, and .claude/agent-memory/ folder with naming conventions and dashboard registration notes.
  • Governance and best practices: Explains gitignore behavior for custom agents, core agent conflicts to avoid, and where to store outputs and memory.
  • Use Case: Onboard a DevOps or customer support assistant by producing a ready-to-use agent markdown, command file, and memory directory.

Quick Start

Tell me the domain, agent name, core responsibilities, personality tone, model choice, and preferred working folder and I will generate the agent file, slash command, and memory folder.

Frequently Asked Questions about create-agent

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

FAQPage Schema
How do I create a custom agent for my Claude workspace?

To create a custom agent for your Claude workspace, you define the domain, name, responsibilities, and model choice. The process generates a system prompt in .claude/agents, a slash command in .claude/commands, and a persistent memory directory under .claude/agent-memory.

What files do I need to set up a workspace agent with a slash command?

Setting up a workspace agent requires a markdown system prompt file in .claude/agents and a command file in .claude/commands. A gitignored memory folder is also created under .claude/agent-memory to store persistent context for the specialized assistant.

Can I assign a specific personality and model to a custom agent?

Yes, you can assign a specific personality tone and model choice to a custom agent. Guided agent definition walks you through selecting these parameters along with the domain, core responsibilities, color, and preferred working folder before generating the files.

What is the best way to onboard a specialized DevOps assistant in Claude?

The best way to onboard a specialized DevOps assistant is through guided agent definition. This produces a ready-to-use agent markdown file, a slash command, and a memory directory without manual file layout or naming guesswork.

Does the agent memory folder get tracked in git?

No, the agent memory folder is not tracked in git. The folder under .claude/agent-memory follows gitignore behavior for custom agents, ensuring that persistent memory and outputs remain local to your workspace environment.

Are there naming convention conflicts to avoid when creating workspace agents?

Yes, you must avoid core agent conflicts when creating workspace agents. Following the specified naming conventions and dashboard registration rules ensures your custom agent files and slash commands do not overwrite existing core agent configurations.