How do I trigger a specific agent?

Agents are triggered based on the 'description' field in their markdown frontmatter. Providing 2-4 concrete examples of user requests and assistant responses helps Claude know when to activate the agent.

Can I restrict the tools available to an agent?

Yes, you can use the 'tools' field in the agent's frontmatter to define an array of specific tools, following the principle of least privilege for better security and focus.

What are the naming conventions for Claude agents?

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

Which AI model should my agent use?

It is generally recommended to use 'inherit' so the agent uses the same model as the parent process, but you can explicitly set it to Sonnet, Opus, or Haiku if specialized capabilities are needed.

What is a Claude Code Agent?

Agents are autonomous subprocesses within Claude Code that handle complex, multi-step tasks independently, as opposed to commands which are user-initiated actions.

Claude Code Agent Architect

Name: Claude Code Agent Architect
Author: chrismcgraw60

bychrismcgraw60

開発者ツール

Provides comprehensive patterns and best practices for creating autonomous agents and specialized sub-processes within Claude Code.

概要

The Agent Development skill is a specialized resource for developers building on the Claude Code platform. It offers standardized templates and architectural guidance for creating autonomous agents capable of executing multi-step workflows. By defining precise triggering conditions, system prompt structures, and tool permissions, this skill ensures that custom agents integrate seamlessly and perform reliably across complex development tasks, moving beyond simple commands to true autonomous operation.

主な機能

System prompt engineering patterns for autonomous behavior
Granular tool access and model configuration guidance
0 GitHub stars
Standardized agent file structure and YAML frontmatter templates
Optimized triggering logic using example-driven descriptions
Validation rules for agent identifiers and metadata

ユースケース

Building specialized code reviewer sub-agents for custom plugins
Architecting autonomous documentation generators within developer workflows
Designing multi-step security analyzers with restricted tool access

概要

主な機能

System prompt engineering patterns for autonomous behavior
Granular tool access and model configuration guidance
0 GitHub stars
Standardized agent file structure and YAML frontmatter templates
Optimized triggering logic using example-driven descriptions
Validation rules for agent identifiers and metadata

ユースケース

Building specialized code reviewer sub-agents for custom plugins
Architecting autonomous documentation generators within developer workflows
Designing multi-step security analyzers with restricted tool access