What is the recommended model for a custom agent?

It is generally recommended to set the model to 'inherit' in the frontmatter, which allows the agent to use the same model currently active in the parent session.

What are the naming conventions for agent identifiers?

Identifiers must be between 3-50 characters, use lowercase letters, numbers, and hyphens only, and must start and end with an alphanumeric character.

How do I define when an agent should be triggered?

Triggering is defined in the agent's markdown frontmatter description field. It must include concrete usage examples with context, user requests, and assistant responses to help Claude identify the right situation.

Can I restrict which tools an agent has access to?

Yes, you can use the 'tools' field in the agent's frontmatter to specify an array of allowed tools, following the principle of least privilege to enhance security and focus.

What is an agent in the context of Claude Code?

An agent is an autonomous subprocess within a Claude Code plugin designed to handle complex, multi-step tasks independently, whereas commands are typically for direct user-initiated actions.

Agent Development for Claude Code

Name: Agent Development for Claude Code
Author: mbarnes-code

bymbarnes-code

開発者ツール

Guides the creation and optimization of autonomous agents within the Claude Code ecosystem through structured prompts and configuration.

概要

The Agent Development skill provides developers with a standardized framework for building, configuring, and testing autonomous subprocesses in Claude Code plugins. It offers expert guidance on crafting YAML frontmatter, defining precise triggering conditions with contextual examples, and designing robust system prompts that ensure agents perform complex, multi-step tasks with high reliability. Whether you are implementing specialized code reviewers or automated security analyzers, this skill ensures your agents follow best practices for tool usage, model selection, and UI integration.

主な機能

AI-assisted agent generation patterns and validation rules
Sophisticated triggering logic using contextual examples and commentary
0 GitHub stars
Standardized markdown-based agent configuration framework
Expert system prompt templates for autonomous behavior
Granular tool permission management for agent security

ユースケース

Building specialized autonomous subagents for complex code analysis
Configuring domain-specific agents with restricted tool access for security
Optimizing agent triggering conditions to reduce false positives in plugin workflows

概要

主な機能

AI-assisted agent generation patterns and validation rules
Sophisticated triggering logic using contextual examples and commentary
0 GitHub stars
Standardized markdown-based agent configuration framework
Expert system prompt templates for autonomous behavior
Granular tool permission management for agent security

ユースケース

Building specialized autonomous subagents for complex code analysis
Configuring domain-specific agents with restricted tool access for security
Optimizing agent triggering conditions to reduce false positives in plugin workflows