Agent Development

Create and validate Claude Code agent files with frontmatter, system prompts, and triggering examples.

Updated Jul 7, 2026
One-click install
npx skills add https://github.com/singhaganesh/Urban-assist --skill agent-development-singhaganesh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Agent Development
Source: https://github.com/singhaganesh/Urban-assist/tree/main/.cursor/skills/agent-development
Command: npx skills add https://github.com/singhaganesh/Urban-assist --skill agent-development-singhaganesh

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Writing Claude Code agents requires precise YAML frontmatter, well-structured system prompts, and triggering examples, and mistakes in any of these cause agents to fail to trigger or behave unpredictably. ## Core Features & Use Cases - Agent File Authoring: Provides the complete agent markdown format covering name, description, model, color, and tools frontmatter fields with validation rules. - System Prompt Patterns: Supplies proven templates for analysis, generation, validation, and orchestration agents with process steps, quality standards, and edge-case handling. - Triggering Example Guidance: Explains how to write example blocks with context, user messages, and commentary so agents trigger reliably. - Validation Script: Includes validate-agent.sh to check frontmatter structure, field formats, and system prompt length. - Use Case: When you need to add a code-reviewer agent to a Claude Code plugin, use this Skill to generate the agent file, write triggering examples, and validate the result before testing. ## Quick Start Create a Claude Code agent that reviews pull requests for security issues and validate the generated agent file.

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 agent file?

Create a markdown file in the plugin's agents/ directory with YAML frontmatter containing name, description, model, and color fields, followed by a system prompt written in second person. The description must include triggering conditions and example blocks.

What frontmatter fields are required for Claude Code agents?

Four fields are required: name (lowercase with hyphens, 3-50 characters), description (triggering conditions with examples), model (inherit, sonnet, opus, or haiku), and color (blue, cyan, green, yellow, magenta, or red). The tools field is optional.

How do I write triggering examples for an agent description?

Each example block needs a Context describing the situation, a user message, an assistant response, and a commentary explaining why the agent triggers. Include 2-4 examples covering explicit requests and proactive triggering with varied phrasings.

How do I validate a Claude Code agent file?

Run the validate-agent.sh script with the agent file path. It checks frontmatter structure, required fields, name format and length, model and color values, and system prompt length, reporting errors and warnings.

When should I restrict an agent's tools?

Restrict tools following least privilege when the agent has a narrow task, such as read-only analysis using Read, Grep, and Glob. Omit the tools field to grant access to all available tools.