🎭 Claude Code Personas

Build stronger products with AI personas that act like your best team members

Get design reviews from an Engineering Manager, strategic insights from a Product Manager, and comprehensive QA analysis using simple prompts. Personas create positive feedback loops that bring diverse perspectives to your work. No servers to manage, no complex setup. Just type, ask, and ship better code.

✨ What This Does

This project provides ready-to-use AI personas that easily integrate with Claude Code and its built-in memory import system. Each persona brings specialized knowledge and perspective to your development projects.

🎯 Starter Personas

📐 Alex, Engineering Manager - Technical architecture, code quality, best practices
💡 Sarah, Product Manager - Requirements analysis, user stories, feature prioritization
📋 Marcus, QA Manager - Testing strategies, quality assurance, bug prevention

💬 How to Use

Once installed, simply ask any persona for help:

claude "Ask the engineering manager to review this API design"
claude "Ask Sarah to help prioritize these features"
claude "Ask the QA manager to design a comprehensive test plan"

Claude Code automatically loads the persona's expertise and provides specialized guidance based on their role and experience.

🎯 Context-Aware Feedback

Personas automatically detect work context and adjust their feedback accordingly:

POC/Spike: Focus on feasibility and learning, not production concerns
MVP: Balance functionality with basic quality gates
Production: Full production readiness review
Bug Fix: Target correctness and regression prevention
Refactoring: Emphasize maintainability without scope creep

The system examines PR titles, commit messages, branch names, and file scope to provide appropriate feedback for each development stage.

🚀 Quick Start

Prerequisites

Claude Code CLI installed
Node.js 18+ and npm
5 minutes of your time

Installation

# 1. Clone and install
git clone https://github.com/jasonhanna/claude-personas.git
cd claude-personas
npm install

# 2. Copy personas to your home directory (~/.claude-agents)
npm run install-templates

# 3. Add personas to User memory (~/.claude/CLAUDE.md)
npm run add-personas

That's it! 🎉 Now you can ask any persona for help in any Claude Code session.

Single Project Setup (alternative)

Want personas available only in specific projects? Add them to project memory instead:

# Add to a specific project
npm run add-personas -- --project ./path/to/your/project

📖 Additional Examples

Engineering Review

claude "Ask Alex to review this database schema for scalability issues"

Leverage the perspective of a seasoned engineering manager to analyze your database design

Feature Planning

claude "Ask the product manager to help write user stories for our new checkout flow"

Sarah applies product strategy expertise to create comprehensive user stories

Quality Assurance

claude "Ask the QA manager to identify edge cases we should test for user authentication"

Marcus leverages 14+ years of QA experience to suggest comprehensive test scenarios

Create Complex Tasks and Parallel Workflow

claude "Create parallel tasks for Alex, Sarah, and Marcus to review this latest pull request, leaving comments on the PR"

Unleash the full power of multi-perspective code review: technical depth, product alignment, and quality assurance in one coordinated AI prompt.

🛠️ Management Commands

Template Management

npm run install-templates     # Install persona templates (~/.claude-agents/personas)
npm run update-templates      # Update existing templates with backup
npm run remove-templates      # Remove templates with backup

Memory Management (User & Project)

npm run add-personas          # Add personas to user memory
npm run update-personas       # Update persona imports in user memory
npm run remove-personas       # Remove personas from user memory

Status and Information

npm run personas-status       # Check installation and configuration
npm run list-personas         # List available personas

Per Project Management

npm run add-personas -- --project /path/to/project
npm run update-personas -- --project /path/to/project
npm run remove-personas -- --project /path/to/project

🎨 Customization

Edit Existing Personas

Once installed, personas are stored as markdown files in ~/.claude-agents/personas/. Edit them to:

Adjust personality and communication style
Add project-specific context
Modify expertise areas
Update decision-making frameworks

# Edit the engineering manager persona
open ~/.claude-agents/personas/engineering-manager.md

TIP: Add Project-Specific Context

Why This Matters: Generic personas provide general guidance, but personas with your project's specific context give targeted, actionable advice that considers your actual architecture, tech stack, and constraints.

The Power of Project Context

When personas understand your project details, they can:

Give relevant architecture advice based on your actual tech stack
Suggest realistic testing strategies for your specific frameworks
Prioritize features considering your actual user base and business model
Identify real risks in your codebase rather than theoretical ones

How to Add Project Context

Let Claude Code analyze your codebase to generate project-specific context:

claude "Analyze this codebase and generate a project context section for an Engineering Manager persona. Include: tech stack, architecture patterns, testing approach, key challenges, performance requirements, and security considerations. Format it as markdown that I can add to the persona file."

For Product Manager context, try:

claude "Analyze this project and create a Product Manager context section including: user personas, business model, key pain points, competitive advantages, success metrics, and release process. Make it specific to this codebase."

For QA Manager context:

claude "Review this codebase and generate a QA context section covering: current test coverage, testing frameworks used, test environments, key areas to test, performance requirements, and what's missing from our quality strategy."

Where to Add Context

Add the generated context to your persona files under the "My Current Context" section:

## My Current Context

### All Projects
- [General standards that apply to all your work]

### Your Project Name
- **Tech Stack**: [Your specific technologies]
- **Architecture**: [Your specific patterns]
- **Testing**: [Your specific approach]
- **Key Challenges**: [Your actual challenges]
- **Performance**: [Your actual requirements]
- **Security**: [Your specific considerations]

Example in Action

Before: "Review this API design for best practices" After: "Review this Express.js API design considering our MongoDB Atlas setup, Auth0 integration, and requirement to handle 1000 concurrent users"

The persona now knows your specific context and gives targeted advice!

Create New Personas

You can create your own personas for any role you need! Simply add new .md files to ~/.claude-agents/personas/ and they'll be automatically discovered.

Step-by-Step Process

Create the persona file

# Create a new persona file
touch ~/.claude-agents/personas/security-engineer.md

Follow the persona format (see template below)
Update your memory imports

npm run update-personas  # Adds new persona to your User memory

Start using your new persona

claude "Ask the security engineer to review this authentication flow"

Persona Template

Create personas following this proven format:

# Security Engineer - Jordan Kim

## About Me
Cybersecurity specialist with 12+ years protecting enterprise applications. I believe security should be built in, not bolted on, with a focus on practical, developer-friendly security practices.

## My Core Responsibilities
- Security architecture design and review
- Threat modeling and risk assessment
- Security code review and vulnerability analysis
- Compliance and regulatory guidance (SOC2, GDPR, etc.)
- Developer security training and best practices
- Incident response and forensics

## My Technical Context
- Focus on OWASP Top 10 and modern attack vectors
- Zero-trust architecture principles
- DevSecOps integration with CI/CD pipelines
- Cloud security (AWS, Azure, GCP) best practices
- Experience with security tools: SAST, DAST, dependency scanning
- Minimum security standards: MFA required, encryption at rest/transit
- Threat modeling using STRIDE methodology

## How I Communicate
- **Tone**: Security-focused, practical, educational
- **Focus**: Risk mitigation, compliance, developer enablement
- **Style**: Clear threat explanations with actionable remediation steps

## My Decision Framework
When evaluating security concerns, I consider:
1. **Risk Assessment**: What's the potential impact and likelihood?
2. **Defense in Depth**: How can we layer multiple protections?
3. **Developer Experience**: How can we make security easy to do right?
4. **Compliance**: What regulatory requirements apply?
5. **Cost vs Risk**: What's the appropriate level of protection?

New Persona Ideas

Here are some additional personas you might consider creating:

👨‍💻 DevOps Engineer - Infrastructure, deployment, monitoring

# Create with: touch ~/.claude-agents/personas/devops-engineer.md
claude "Ask the DevOps engineer to design a CI/CD pipeline for this microservice"

🎨 UX Designer - User experience, interface design, usability

# Create with: touch ~/.claude-agents/personas/ux-designer.md  
claude "Ask the UX designer to improve the user flow for our onboarding"

📊 Data Scientist - Analytics, ML/AI, data insights

# Create with: touch ~/.claude-agents/personas/data-scientist.md
claude "Ask the data scientist to recommend metrics for this feature"

🏗️ Solutions Architect - System design, scalability, integrations

# Create with: touch ~/.claude-agents/personas/solutions-architect.md
claude "Ask the solutions architect to design a scalable event processing system"

🔒 Compliance Officer - Regulatory requirements, audit preparation

# Create with: touch ~/.claude-agents/personas/compliance-officer.md
claude "Ask the compliance officer to review our data retention policies"

Best Practices for Customizing Personas

✅ Do:

Give personas specific expertise areas and years of experience
Include their decision-making frameworks and methodologies
Add relevant technical context for your industry/domain
Define clear communication style and tone
Include project memories section for context building

❌ Avoid:

Generic or overly broad expertise claims
Conflicting responsibilities between personas
Too many personas (start with 3-5 core roles)
Copying existing persona content without customization

🔧 How It Works

Simple Architecture

Your Claude Code Session
    ↓ "Ask the engineering manager..."
Claude Memory System
    ↓ Loads: @~/.claude-agents/personas/engineering-manager.md
Persona Context + Your Project
    ↓ 
Expert Response with Role-Specific Guidance

Memory Import System

Uses Claude Code's native @path/to/file memory imports
No servers, no complex infrastructure
Personas load automatically when referenced
Works with all existing Claude Code features

Context-Aware Feedback Protocol

When personas are added to your memory, they include a comprehensive framework that:

Detects work context from PR titles, commit messages, branch names, and issue labels
Calibrates feedback appropriately (e.g., no production concerns for POCs)
Provides stage-specific guidance for different development phases
Asks for clarification when context is unclear ("What stage is this work in?")

File Structure

~/.claude-agents/personas/     # Persona definitions
├── engineering-manager.md     # Alex's expertise and context
├── product-manager.md         # Sarah's expertise and context
└── qa-manager.md             # Marcus's expertise and context

~/.claude/CLAUDE.md           # User memory (global)
OR
./CLAUDE.md                   # Project memory (local)

templates/                    # Template files
└── persona-section.md       # Context-aware feedback framework

🧪 Testing

This project includes comprehensive tests to ensure reliability:

npm test                      # Run all tests
npm run test:coverage        # Run with coverage report
npm test -- persona-scripts  # Run persona-specific tests

🤝 Contributing

Adding New Personas

Create persona markdown file following existing format
Add appropriate emoji icon in scripts
Test with npm run install-templates && npm run update-personas
Submit PR with new persona

Improving Existing Personas

Edit persona files in personas/ directory
Test changes with npm run update-templates
Submit PR with improvements

Development Setup

git clone https://github.com/jasonhanna/claude-personas.git
cd claude-personas
npm install
npm test                      # Ensure tests pass

📚 Documentation

Simplified Design Approach - Technical architecture overview
Changelog - Version history and updates

💡 Use Cases

Code Reviews

Architecture analysis and recommendations
Security vulnerability identification
Performance optimization suggestions
Code quality and maintainability feedback

Project Planning

Feature prioritization and roadmap planning
User story creation and refinement
Technical feasibility assessment
Risk analysis and mitigation strategies

Quality Assurance

Test strategy development
Edge case identification
Quality metrics definition
Bug prevention strategies

Team Collaboration

Cross-functional perspective gathering
Decision-making support
Knowledge sharing and mentoring
Best practice enforcement

🌟 Why Use Personas?

Instant Expertise

Get specialized guidance without hiring consultants or waiting for team members.

Consistent Quality

Each persona maintains consistent standards and approaches across all interactions.

Comprehensive Coverage

Technical, product, and quality perspectives ensure well-rounded solutions.

Learning Opportunity

Learn best practices and decision-making frameworks from experienced professionals.

Always Available

24/7 access to expert guidance for any development challenge.

📄 License

This project is licensed under the MIT License.

🙏 Acknowledgments

Built for Claude Code memory import system
Inspired by the need for specialized AI assistance in software development
Designed for simplicity and immediate productivity gains

Ready to enhance your development workflow with AI expertise?

🚀 Run npm run install-templates && npm run add-personas to get started!

Questions? Problems? Feedback? Open an issue

Name		Name	Last commit message	Last commit date
docs		docs
images		images
personas		personas
scripts		scripts
templates		templates
tests		tests
.gitignore		.gitignore
CHANGELOG.md		CHANGELOG.md
CLAUDE.md		CLAUDE.md
LICENSE.md		LICENSE.md
README.md		README.md
package-lock.json		package-lock.json
package.json		package.json
tsconfig.json		tsconfig.json

Latest commit

History

Folders and files

Repository files navigation

🎭 Claude Code Personas

✨ What This Does

🎯 Starter Personas

💬 How to Use

🎯 Context-Aware Feedback

🚀 Quick Start

Prerequisites

Installation

Single Project Setup (alternative)

📖 Additional Examples

Engineering Review

Feature Planning

Quality Assurance

Create Complex Tasks and Parallel Workflow

🛠️ Management Commands

Template Management

Memory Management (User & Project)

Status and Information

Per Project Management

🎨 Customization

Edit Existing Personas

TIP: Add Project-Specific Context

The Power of Project Context

How to Add Project Context

Where to Add Context

Example in Action

Create New Personas

Step-by-Step Process

Persona Template

New Persona Ideas

Best Practices for Customizing Personas

🔧 How It Works

Simple Architecture

Memory Import System

Context-Aware Feedback Protocol

File Structure

🧪 Testing

🤝 Contributing

Adding New Personas

Improving Existing Personas

Development Setup

📚 Documentation

💡 Use Cases

Code Reviews

Project Planning

Quality Assurance

Team Collaboration

🌟 Why Use Personas?

Instant Expertise

Consistent Quality

Comprehensive Coverage

Learning Opportunity

Always Available

📄 License

🙏 Acknowledgments

About

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages