Agent Development Framework FAQs

Question 1

What is an agent in the context of Claude Code?

Accepted Answer

An agent is an autonomous subprocess designed to handle complex, multi-step tasks independently using a specific system prompt, color identity, and restricted toolset.

Question 2

What are the naming rules for new agents?

Accepted Answer

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

Question 3

Which AI model should my agent use?

Accepted Answer

It is generally recommended to set the model to 'inherit', which allows the agent to use the same model as the parent process, though you can specify Opus, Sonnet, or Haiku for specific needs.

Question 4

Can I restrict what tools an agent can use?

Accepted Answer

Yes, the 'tools' field in the YAML frontmatter allows you to define an array of allowed tools, following the principle of least privilege to enhance safety and focus.

Question 5

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

Accepted Answer

Triggering is controlled by the description field in the frontmatter. You should provide 2-4 concrete examples using XML-style tags to show Claude exactly when the agent is appropriate.

Agent Development Framework

主要功能

使用场景

Agent Development Framework

主要功能

使用场景