What are the required fields for an agent's frontmatter?

Every agent requires a unique name (identifier), a description with triggering examples, a specified model (usually 'inherit'), and a color for visual identification.

What is a Claude Code Agent?

Agents are autonomous subprocesses within Claude Code that handle complex, multi-step tasks independently using specific system prompts and tools.

How do I ensure my agent triggers at the right time?

Include 2-4 concrete examples in the description field using context, user, and assistant tags to provide the model with clear patterns for when to invoke the agent.

When should I use an agent instead of a command?

Use agents for autonomous, multi-step work that requires a specific persona; use commands for direct, user-initiated actions like simple file operations.

How do I control which tools an agent can access?

You can use the 'tools' field in the YAML frontmatter to list specific tools like 'Read', 'Write', or 'Bash', restricting the agent to the minimum access needed.

Claude Code Agent Developer

Name: Claude Code Agent Developer
Author: michelabboud

bymichelabboud

•

开发者工具

Provides comprehensive guidance and templates for building, structuring, and optimizing autonomous agents within Claude Code plugins.

关于

The Agent Development skill empowers developers to create sophisticated autonomous subprocesses for Claude Code. It provides a standardized framework for defining agent identifiers, triggering conditions through detailed examples, and crafting effective system prompts. Whether you are building from scratch or using AI-assisted generation, this skill ensures your agents are well-structured, properly validated, and equipped with the right tools and models to handle complex, multi-step tasks independently.

主要功能

Best practices for system prompt engineering and persona design
1 GitHub stars
Step-by-step workflows for AI-assisted or manual agent creation
Standardized YAML frontmatter templates for agent configuration
Advanced triggering logic with example-based description fields
Validation rules for agent identifiers, colors, and tool sets

使用场景

Creating specialized subagents for code review, testing, or API documentation
Optimizing agent triggering conditions to improve autonomous task performance
Developing a structured plugin ecosystem with namespaced agents

关于

主要功能

Best practices for system prompt engineering and persona design
1 GitHub stars
Step-by-step workflows for AI-assisted or manual agent creation
Standardized YAML frontmatter templates for agent configuration
Advanced triggering logic with example-based description fields
Validation rules for agent identifiers, colors, and tool sets

使用场景

Creating specialized subagents for code review, testing, or API documentation
Optimizing agent triggering conditions to improve autonomous task performance
Developing a structured plugin ecosystem with namespaced agents