Agent Development

Design and validate Claude Code agent specifications with frontmatter and system prompts.

2|1|Updated Apr 9, 2026
One-click install
npx skills add https://github.com/timgranlundmarsden/claude-agent-flow --skill agent-development-timgranlundmarsden
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Agent Development
Source: https://github.com/timgranlundmarsden/claude-agent-flow/tree/main/skills/agent-development
Command: npx skills add https://github.com/timgranlundmarsden/claude-agent-flow --skill agent-development-timgranlundmarsden

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps you design, structure, and validate Claude Code agents so they trigger reliably, behave consistently, and follow the right development patterns.

Core Features & Use Cases

  • Agent Design Guidance: Define a clear agent role, responsibilities, and operating boundaries for autonomous work.
  • Triggering and Frontmatter: Write discovery-friendly descriptions with examples and the required metadata fields for agent files.
  • Prompt and Validation Best Practices: Shape effective system prompts, choose appropriate tools and colors, and validate the final agent file before use.
  • Use Case: A developer wants to add a code review agent, a test generator, or a documentation helper to a Claude Code plugin and needs a reliable template that works with the expected file structure.

Quick Start

Use the agent-development skill to create a new Claude Code agent file with the correct frontmatter, triggering examples, and system prompt.

Frequently Asked Questions about Agent Development

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

FAQPage Schema
How do I create a Claude Code subagent for autonomous code review?

To create a Claude Code subagent, define a clear agent role, write required frontmatter fields, add triggering examples, and structure a second-person system prompt to validate the agent file before use.

What frontmatter fields are required for Claude Code agent files?

Claude Code agent files require specific frontmatter fields, including discovery-friendly descriptions, triggering examples, optional tool restrictions, and validation-ready markdown structure to ensure reliable agent behavior.

How do I write system prompts for Claude Code subagents?

Write system prompts for Claude Code subagents in the second person, clearly defining the agent's role, responsibilities, and operating boundaries to ensure consistent autonomous behavior.

Can I restrict which tools a Claude Code subagent can access?

Yes, you can apply optional tool restrictions when designing Claude Code agent specifications, limiting the subagent's capabilities to only the tools needed for its specific workflow.

Why are my Claude Code subagents not triggering reliably?

Subagents fail to trigger reliably when their agent files lack discovery-friendly descriptions, proper frontmatter fields, or sufficient triggering examples needed for Claude Code to match the agent to the task.

What is the best way to structure a Claude Code plugin workflow?

The best way to structure a Claude Code plugin workflow is to design autonomous subagents with clear operating boundaries, validated system prompts, and proper frontmatter for tasks like test generation or documentation.