What problem does it solve?
This Skill simplifies and standardizes the creation of Claude Code agents, eliminating manual configuration and ensuring consistent, discoverable, and effective AI agents. It saves time by guiding you through best practices for agent design and deployment.
Core Features & Use Cases
- Declarative Agent Definition: Create agents with YAML frontmatter for clear names, descriptions, scoped tool permissions, and model selection.
- Optimized Discovery: Craft compelling descriptions with "Use when" clauses and proactive triggers to maximize agent discoverability and auto-invocation.
- Structured System Prompts: Design effective system prompts with clear role lines, numbered sections, and constraint markers to guide agent behavior.
- Workflow-Driven Creation: Follow a 5-phase workflow (Understand, Acquire, Research, Author, Validate) to ensure research-informed and compliant agent creation.
- Use Case: When you need a specialized agent for code review, use this Skill to define its
name, description, tools (e.g., Read, Glob), model (e.g., sonnet), and system prompt, ensuring it's ready for deployment.
Quick Start
Use the agent-builder skill to create a new code-reviewer agent with Read and Glob tools, using the sonnet model, and a system prompt focused on identifying security vulnerabilities.