Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

New Agent

ASecurity

Guide through creating a new purpose-built agent for specialized tasks with proper configuration

8 stars
0 votes
0 copies
0 views
Added 9/20/2026
code-qualitygojavatestingdebugginggitci/cdperformancedocumentation

Works with

claude code

Security Analysis

A100/100

Scanned 9/20/2026

Install to Claude Code

$npx -y skills add tstapler/dotfiles --skill new-agent --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of New Agent?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for New Agent
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/tstapler-new-agent/badge)](https://www.skillsdirectory.com/skills/tstapler-new-agent)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
description: Guide through creating a new purpose-built agent for specialized tasks with proper configuration
---

# Creating New Claude Code Agent: `${1:-example-agent}`

I'll help you create a new purpose-built agent for specialized tasks, following your established agent configuration patterns.

## Agent Configuration

**Agent Name**: `${1:-example-agent}` (kebab-case format)
**Specialization Domain**: `${2:-general-purpose}`
**Model Preference**: `${3:-sonnet}` (Options: sonnet, opus, haiku - prefer sonnet for speed)

## Agent Creation Process

### Step 1: Define Agent Scope and Expertise

**Core Questions to Answer:**
- What specific domain expertise does this agent provide?
- When should users invoke this agent vs. general Claude?
- What tools are essential for this agent's function?
- What knowledge areas require deep, specialized context?

### Step 2: Analyze Context Efficiency Potential

**High Priority Candidates (Create Agent):**
- Complex, multi-step workflows requiring specialized knowledge
- Domain-specific expertise (JIRA, testing, architecture, etc.)
- Repetitive tasks with established patterns and best practices
- Workflows requiring consistent methodology and quality standards
- Tasks that benefit from maintained context and state

**Low Priority Candidates (Keep as Command):**
- Simple, procedural workflows
- One-time or infrequent tasks
- Basic formatting or templating operations
- Tasks requiring minimal specialized knowledge

### Step 3: Create Agent File Structure

**File Location**: `/Users/tylerstapler/.claude/agents/${1:-example-agent}.md`

**Required YAML Frontmatter:**
```yaml
---
name: ${1:-example-agent}
description: Use this agent when [specific trigger conditions]. This agent should be invoked [when to use it based on context and examples].

Examples:
- <example>
  Context: [Situation description]
  user: "[Example user request]"
  assistant: "[Response showing agent invocation]"
  <commentary>
  [Explanation of why this agent is appropriate for this scenario]
  </commentary>
  </example>

tools: [list of required tools or * for all tools]
model: ${3:-sonnet}
---
```

### Step 4: Define Agent Expertise and Methodology

**Core Structure:**
1. **Mission Statement**: Clear purpose and role definition
2. **Key Expertise Areas**: Domain-specific knowledge and competencies
3. **Methodology/Process**: Step-by-step approach for handling requests
4. **Quality Standards**: Non-negotiable requirements and success criteria
5. **Best Practices**: Domain-specific patterns and recommendations

### Step 5: Implementation Template

Based on your inputs, here's the agent template:

```markdown
---
name: ${1:-example-agent}
description: Use this agent when you need specialized ${2:-general-purpose} expertise. This agent should be invoked when [specific conditions that warrant specialized knowledge over general Claude capabilities].

Examples:
- <example>
  Context: User needs ${2:-general-purpose} assistance.
  user: "I need help with [specific ${2:-general-purpose} task]"
  assistant: "I'll use the ${1:-example-agent} agent to provide specialized ${2:-general-purpose} expertise"
  <commentary>
  Since this requires ${2:-general-purpose} domain knowledge, the ${1:-example-agent} agent is the appropriate choice.
  </commentary>
  </example>

tools: *
model: ${3:-sonnet}
---

You are a ${2:-general-purpose} specialist with deep expertise in [specific domain areas]. Your role is to provide expert-level assistance while maintaining [specific quality standards].

## Core Mission

[Define the primary purpose and value proposition of this agent]

## Key Expertise Areas

### **[Domain Area 1]**
- [Specific competency]
- [Related skill or knowledge]
- [Best practices and patterns]

### **[Domain Area 2]**  
- [Specific competency]
- [Related skill or knowledge]
- [Best practices and patterns]

## Methodology

### **Phase 1: [Process Step]**
[Description of what happens in this phase]

### **Phase 2: [Process Step]**
[Description of what happens in this phase]

### **Phase 3: [Process Step]**
[Description of what happens in this phase]

## Quality Standards

You maintain these non-negotiable standards:
- [Standard 1]: [Description and rationale]
- [Standard 2]: [Description and rationale]
- [Standard 3]: [Description and rationale]

## Professional Principles

- [Principle 1]: [How this guides your work]
- [Principle 2]: [How this guides your work]
- [Principle 3]: [How this guides your work]

Remember: [Key reminder about the agent's core purpose and value]
```

## Current Agent Ecosystem

**Existing Specialized Agents:**
- `pr-reviewer`: Code review and software engineering best practices
- `java-test-debugger`: Java testing framework debugging and fixes
- `github-debugger`: GitHub Actions and CI/CD troubleshooting
- `jira-project-manager`: FBG JIRA and project management workflows
- `knowledge-synthesis`: Zettelkasten research and knowledge integration

## Agent Design Principles

**Context Efficiency Guidelines:**
- **High Specialization**: Focus on narrow, deep expertise vs. broad knowledge
- **Consistent Methodology**: Establish repeatable processes and quality standards  
- **Tool Optimization**: Include only tools essential for the domain
- **Model Selection**: Prefer sonnet for speed and efficiency; use opus only for extremely complex reasoning tasks; use haiku for simple tasks
- **Clear Boundaries**: Define exactly when to use this agent vs. general Claude

**Usage Pattern Design:**
- **Trigger Conditions**: Make it obvious when to invoke this agent
- **Example Scenarios**: Provide concrete examples with commentary
- **Value Proposition**: Clearly articulate why specialization matters
- **Quality Standards**: Define what makes this agent's output superior

## Implementation Steps

1. **Create Agent File**: `/Users/tylerstapler/.claude/agents/${1:-example-agent}.md`
2. **Define Specialization**: Focus on specific domain expertise
3. **Configure Tools**: Include essential tools for the domain
4. **Set Quality Standards**: Define non-negotiable requirements
5. **Test and Refine**: Validate agent performance vs. general Claude
6. **Document Usage**: Update any relevant workflow documentation

Would you like me to create this `${1:-example-agent}` agent now based on your specifications? Please confirm:
- **Agent Name**: `${1:-example-agent}`
- **Specialization Domain**: `${2:-general-purpose}`
- **Model**: `${3:-sonnet}` (prefer sonnet for better performance)
- **Specific Expertise Areas**: [Please specify the key domains]
- **Essential Tools**: [Please specify required tools]

Attribution

tstaplertstapler
View sourceMore from tstapler →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Caveman Commit

Ultra-compressed commit message generator. Cuts noise from commit messages while preserving intent and reasoning. Conventional Commits format. Subject ≤50 chars, body only when "why" isn't obvious. Use when user says "write a commit", "commit message", "generate commit", "/commit", or invokes /caveman-commit. Auto-triggers when staging changes.

1023331 votes

Caveman Review

Ultra-compressed code review comments. Cuts noise from PR feedback while preserving the actionable signal. Each comment is one line: location, problem, fix. Use when user says "review this PR", "code review", "review the diff", "/review", or invokes /caveman-review. Auto-triggers when reviewing pull requests.

1023331 votes

Verification Loop

一个全面的 Claude Code 会话验证系统。

2456590 votes

Springboot Verification

Verification loop for Spring Boot projects: build, static analysis, tests with coverage, security scans, and diff review before release or PR.

2456590 votes

Django Verification

Verification loop for Django projects: migrations, linting, tests with coverage, security scans, and deployment readiness checks before release or PR.

2456590 votes
View all in code-quality →