Agent Development
Author Claude Code plugin agents with correct frontmatter, triggers, and system prompts instead of misusing commands or skills.
Overview
Agent Development is an agent skill for the Build phase that teaches solo builders how to structure, trigger, and prompt autonomous Claude Code plugin agents.
Install
npx skills add https://github.com/anthropics/claude-plugins-official --skill agent-developmentWhat is this skill?
- Defines the full agent markdown + YAML frontmatter format (name, description, model, color, tools).
- Separates agents (autonomous multi-step), commands (user-initiated), and skills (automatic hooks) with selection criteri
- Teaches prose-style triggering in description plus worked scenarios in the agent body to reduce over- and under-triggeri
- Provides a five-part system prompt skeleton: role, responsibilities, approach, output, and constraints.
- Includes a testing checklist: trigger phrases, over-triggering, under-triggering, and output quality.
- Five-part system prompt structure: role, responsibilities, approach, output, constraints
- Agent file uses YAML frontmatter fields including name, description, model, color, and optional tools array
- Testing checklist covers trigger phrases, over-triggering, under-triggering, and output quality
Adoption & trust: 3.4k installs on skills.sh; 29.6k GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You need a reliable autonomous subprocess in a Claude Code plugin but are unsure how to format the agent file, write triggers, or distinguish agents from commands and skills.
Who is it for?
Indie builders and small teams authoring or refactoring Claude Code plugin agents who want Anthropic-aligned structure before sharing plugins.
Skip if: Builders who only need a one-off slash command, a packaged SKILL.md workflow, or who are not working inside Claude Code plugins.
When should I use this skill?
User asks to create an agent, add an agent, write a subagent, configure agent frontmatter, when-to-use descriptions, agent examples, tools, colors, autonomous agents, or agent development best practices for Claude Code p
What do I get? / Deliverables
You ship a tested agent markdown file with clear YAML frontmatter, scenario-based triggers, and a structured system prompt that invokes on the right tasks without constant misfires.
- Agent markdown file with complete YAML frontmatter
- Prose-based trigger description and when-to-invoke scenarios
- Structured system prompt aligned to agent responsibilities
Recommended Skills
Journey fit
The skill exclusively documents how to create autonomous plugin agents inside Claude Code, which is core product and agent-infrastructure work during Build. Content covers agent markdown files, YAML frontmatter, tool lists, triggering descriptions, and testing agents—squarely agent-tooling for plugins.
How it compares
Use for autonomous multi-step plugin subprocesses, not slash-command shortcuts or generic SKILL.md procedural packages.
Common Questions / FAQ
Who is agent-development for?
It is for solo builders and plugin authors who extend Claude Code with autonomous agents and need consistent frontmatter, triggers, and prompts.
When should I use agent-development?
Use it during Build when you are asked to create or add an agent, write subagent frontmatter, fix triggering conditions, tune agent tools and colors, or align agent examples with Claude Code plugin conventions.
Is agent-development safe to install?
Treat it like any third-party skill: review the Security Audits panel on this Prism page and your org policy before enabling it in production plugin repos.
SKILL.md
READMESKILL.md - Agent Development
# Agent Development for Claude Code Plugins ## Overview Agents are autonomous subprocesses that handle complex, multi-step tasks independently. Understanding agent structure, triggering conditions, and system prompt design enables creating powerful autonomous capabilities. **Key concepts:** - Agents are FOR autonomous work, commands are FOR user-initiated actions - Markdown file format with YAML frontmatter - Triggering via description field with examples - System prompt defines agent behavior - Model and color customization ## Agent File Structure ### Complete Format ```markdown --- name: agent-identifier description: Use this agent when [triggering conditions]. Typical triggers include [scenario 1 in prose], [scenario 2 in prose], and [scenario 3 in prose]. See "When to invoke" in the agent body for worked scenarios. model: inherit color: blue tools: ["Read", "Write", "Grep"] --- You are [agent role description]... ## When to invoke [Two to four representative scenarios written as prose, e.g.:] - **[Scenario name].** [What the situation looks like and what the agent should do.] - **[Scenario name].** [Same.] **Your Core Responsibilities:** 1. [Responsibility 1] 2. [Responsibility 2] **Analysis Process:** [Step-by-step workflow] **Output Format:** [What to return] ``` ## Frontmatter Fields ### name (required) Agent identifier used for namespacing and invocation. **Format:** lowercase, numbers, hyphens only **Length:** 3-50 characters **Pattern:** Must start and end with alphanumeric **Good examples:** - `code-reviewer` - `test-generator` - `api-docs-writer` - `security-analyzer` **Bad examples:** - `helper` (too generic) - `-agent-` (starts/ends with hyphen) - `my_agent` (underscores not allowed) - `ag` (too short, < 3 chars) ### description (required) Defines when Claude should trigger this agent. **This is the most critical field** — it is loaded into context whenever the agent is registered, so the harness can decide when to dispatch. **Must include:** 1. Triggering conditions ("Use this agent when...") 2. A short prose summary of the typical trigger scenarios 3. A pointer to a "When to invoke" section in the agent body for the detailed worked scenarios **Format:** ``` Use this agent when [conditions]. Typical triggers include [scenario 1 in prose], [scenario 2 in prose], and [scenario 3 in prose]. See "When to invoke" in the agent body for worked scenarios. ``` **Best practices:** - Name 2-4 trigger scenarios in the prose summary - Cover both proactive (assistant invokes itself) and reactive (user requests) triggering - Cover different phrasings of the same intent - Be specific about when NOT to use the agent - Put detailed scenarios in the body under "When to invoke" as a bullet list of prose descriptions ### model (required) Which model the agent should use. **Options:** - `inherit` - Use same model as parent (recommended) - `sonnet` - Claude Sonnet (balanced) - `opus` - Claude Opus (most capable, expensive) - `haiku` - Claude Haiku (fast, cheap) **Recommendation:** Use `inherit` unless agent needs specific model capabilities. ### color (required) Visual identifier for agent in UI. **Options:** `blue`, `cyan`, `green`, `yellow`, `magenta`, `red` **Guidelines:** - Choose distinct colors for different agents in same plugin - Use consistent colors for similar agent types - Blue/cyan: Analysis, review - Green: Success-oriented tasks - Yellow: Caution, validation - Red: Critical, security - Magenta: Creative, generation ### tools (optional) Restrict agent to specific tools. **Form