The Principal Agent Collection

Welcome to a curated collection of 50 principal-level AI agents, designed to function as a world-class, multi-disciplinary team of virtual experts. Each agent embodies the knowledge, principles, and strategic thinking of a senior leader in their respective field.

This is more than a set of tools; it's a team of seasoned professionals ready to tackle your most complex challenges, from architectural design and AI implementation to security audits and go-to-market strategy.

Guiding Philosophy

These agents were built on a foundation of core principles to ensure they provide strategic, secure, and actionable guidance:

• Principal-Level Expertise: Each agent operates with the strategic oversight of a seasoned professional, focusing on long-term value, scalability, and maintainability.
• Security by Design: Security is a foundational, non-negotiable requirement integrated into every relevant agent's core logic.
• Actionable Deliverables: Agents are designed to produce clear, practical, and high-value outputs, such as architectural diagrams, strategic plans, and production-ready code.
• Modern Best Practices: The agents' knowledge bases are centered on current, industry-leading best practices, tools, and architectural patterns.
• Pragmatism over Dogma: While based on strong principles, the agents favor practical solutions that solve real-world problems over rigid, theoretical purity.

Table of Contents

• The Principal Agent Roster
• Integration with Claude Code
• Usage & Best Practices
• Example Workflows
• Contributing
• License

🤖 The Principal Agent Roster

Here is the complete roster of your 50-agent expert team, organized by their core domain.

1. Core Engineering & Architecture

Your virtual CTO office and senior engineering leaders for designing robust, scalable systems.

• backend-architect: Designs scalable, cloud-native backend systems, focusing on microservices, API design, and data modeling.
• cloud-architect: Leads cloud infrastructure strategy across AWS, GCP, and Azure, optimizing for cost, scalability, and resilience.
• graphql-architect: A specialist in designing and scaling federated GraphQL APIs for enterprise applications.
• legacy-modernizer: An expert in strategically refactoring legacy systems, migrating to modern frameworks, and reducing technical debt.
• architect-review: The guardian of architectural integrity; reviews changes for alignment with long-term vision and best practices.

2. Language & Platform Specialists

Distinguished engineers with deep, idiomatic expertise in specific programming languages and platforms.

• python-pro: Master of idiomatic Python, focusing on performance, concurrency, and robust application structure.
• javascript-pro: Expert in the full JavaScript/TypeScript ecosystem, building large-scale, resilient applications for browser and Node.js.
• golang-pro: Specialist in building high-performance, concurrent systems using idiomatic Go.
• rust-pro: Master of building high-assurance software that is safe, concurrent, and fast by default.
• sql-pro: Principal Database Engineer for designing schemas, optimizing complex queries, and ensuring data integrity.
• c-pro: Expert in high-performance C, focusing on systems programming, memory management, and POSIX APIs.
• cpp-pro: Master of modern C++, RAII, and high-performance computing.
• regex-expert: A dedicated specialist for crafting and debugging complex regular expressions.

3. AI, Machine Learning & Data

Your dedicated AI and Data leadership for building intelligent systems and data-driven insights.

• ai-engineer: Principal AI Engineer for designing and building production-grade, scalable AI systems and advanced RAG pipelines.
• ml-engineer: Principal ML Engineer managing the end-to-end lifecycle of machine learning models, from feature engineering to production monitoring.
• data-engineer: Principal Data Engineer architecting modern, scalable data platforms that are automated, self-service, and cost-efficient.
• data-scientist: Principal Data Scientist focused on hypothesis-driven analysis, causal inference, and predictive modeling.
• mlops-engineer: Manages the complete MLOps lifecycle, ensuring models are deployed, monitored, and governed efficiently.
• prompt-engineer: Principal Prompt Engineer architecting advanced prompt systems, implementing automated evaluation, and ensuring prompt security.
• quant-analyst: Builds quantitative models, backtests trading strategies, and performs rigorous financial data analysis.

4. Infrastructure, Operations & Tooling

Your senior team for building and running world-class infrastructure and empowering developers.

• deployment-engineer: Manages CI/CD pipelines, containerization, and cloud-native deployments.
• devops-troubleshooter: Expert in diagnosing and resolving complex production issues across the stack.
• dx-optimizer: Principal Developer Experience Engineer focused on creating a world-class internal developer platform and optimizing workflows.
• terraform-specialist: Infrastructure as Code expert for writing advanced Terraform modules and managing complex state.
• database-admin: Manages high-availability database clusters, replication, backups, and performance tuning.
• network-engineer: Designs and debugs complex network architectures, from VPCs and load balancers to DNS.
• context-manager: A utility agent that helps manage and synthesize context for other agents in complex tasks.

5. Quality, Security & Reliability

Your dedicated experts for ensuring your systems are secure, reliable, and built to the highest quality standards.

• security-auditor: A security professional who performs vulnerability assessments, threat modeling, and security architecture reviews.
• performance-engineer: A specialist in diagnosing and resolving performance bottlenecks in applications and infrastructure.
• test-automator: Designs and implements comprehensive, multi-layered testing strategies (unit, integration, E2E).
• incident-responder: Manages production incidents with a calm, systematic approach to minimize impact and identify root causes.
• accessibility-auditor: Ensures applications are compliant with WCAG and accessible to all users.
• code-reviewer: Provides expert, in-depth code reviews focused on quality, maintainability, and best practices.
• debugger: A systematic bug hunter who can track down the root cause of complex software defects.
• error-detective: Specializes in analyzing logs and error patterns to find the source of elusive issues.
• database-optimizer: A specialist focused purely on query optimization and advanced indexing strategies.

6. Product, Design & Documentation

Your product leadership team for creating world-class user experiences and documentation.

• ux-ui-designer: Designs intuitive, user-centric interfaces and experiences, providing wireframes, mockups, and design systems.
• technical-writer: Lead Technical Writer & Documentation Architect who designs and implements a comprehensive documentation strategy.
• api-documenter: A specialist for generating and maintaining clear, comprehensive API documentation using standards like OpenAPI.
• changelog-generator: Automatically creates clear, well-formatted changelogs from commit histories.
• frontend-developer: Builds modern, responsive user interfaces with a focus on component architecture and state management.
• mobile-developer: Develops cross-platform mobile applications using frameworks like React Native and Flutter.

7. Business, Marketing & Legal

Your go-to-market and strategic advisory team for driving business growth and ensuring compliance.

• content-marketer: Principal Content Strategist who develops data-driven content programs to build brand authority and drive growth.
• sales-automator: Principal Revenue Operations Architect who designs the scalable technology and process backbone for the GTM team.
• legal-advisor: Principal Corporate Counsel providing strategic legal guidance on technology transactions, IP, and data privacy.
• business-analyst: Analyzes business requirements, models financial projections, and defines key performance indicators.
• payment-integration: A specialist for securely integrating payment processors like Stripe and PayPal.
• risk-manager: Identifies, assesses, and mitigates business and financial risks.
• search-specialist: Conducts deep market research, competitive analysis, and fact-checking.
• customer-support: Manages customer communications, develops support workflows, and creates help documentation.

Integration with Claude Code

This agent collection is designed to be seamlessly integrated into AI-powered development environments like Claude Code. By loading these agent definitions, you transform your coding assistant from a general-purpose tool into a specialized team of experts.

How it Works:

When you provide this collection to Claude Code, it indexes the 50 expert personas defined in the markdown files. Each file acts as a "prompt-sheet" that gives the AI a specific role, set of skills, and strategic principles.

Once loaded, Claude Code's routing system can delegate your requests to the most appropriate expert. For example, a question about database performance will be automatically routed to the @database-optimizer, while a request to design a new feature will bring in the @backend-architect and @ux-ui-designer.

How to Use:

1. Load the Agents: The primary step is to make your Claude Code environment aware of this agent directory. This is typically done in the settings or configuration panel of the tool, where you can specify a path to a local folder containing agent definitions. Point it to the root of this repository.
2. Invoke Agents in Chat: Once loaded, you can interact with the agents directly in your chat with Claude Code:
   • Automatic Delegation: For general tasks, simply state your goal. Claude's internal routing will pick the best agent for the job.

     "My application's login page is slow. Please investigate and fix it." (This will likely activate the @performance-engineer, @devops-troubleshooter, and relevant language pro).

   • Explicit Delegation: For specific tasks, you can call an agent by its handle (@agent-name) to ensure the right expert is on the case.

     "@security-auditor, please perform a threat model for our new user authentication service."

   • Multi-Agent Collaboration: Assemble a team for complex projects by mentioning multiple agents.

     "@ai-engineer and @legal-advisor, I need to build a RAG pipeline that uses customer data. Design a system that is both effective and privacy-compliant."

By following this approach, you leverage the full power of this specialized agent collection, ensuring that every task is handled by an expert with the right strategic mindset.

Usage & Best Practices

Think of this not as invoking a tool, but as delegating a task to a senior member of your team. The more context and clarity you provide, the better the outcome.

• Trust the System: For general tasks, describe your goal and let the system delegate to the appropriate expert.
• Delegate Explicitly: For specific needs, call an agent by name: "@security-auditor Please review this authentication flow for vulnerabilities."
• Assemble a Task Force: Combine multiple experts for complex, multi-disciplinary projects: "@cloud-architect and @backend-architect, design a resilient, multi-region architecture for our new service."
• Empower with Context: Provide agents with the necessary background, constraints, and objectives, just as you would with a human expert.

Example Workflows

Launching a New AI-Powered Feature

A strategic, multi-agent collaboration to take a feature from concept to production.

1. @business-analyst: Defines the feature's goals, KPIs, and target user profile.
2. @ai-engineer: Designs the core AI system architecture, including model selection and data pipelines.
3. @backend-architect: Designs the APIs and microservices to support the AI feature.
4. @legal-advisor: Reviews the data usage and model behavior for privacy and compliance risks.
5. @security-auditor: Performs a threat model of the new architecture.
6. @deployment-engineer: Plans the CI/CD and deployment strategy for the new services and models.
7. @content-marketer: Develops a go-to-market content plan for the new feature.

Responding to a Production Performance Degradation

A coordinated incident response by a specialized team.

1. @incident-responder: Takes command of the incident, establishing communication channels and a systematic investigation process.
2. @devops-troubleshooter: Analyzes monitoring dashboards, logs, and traces to isolate the area of degradation.
3. @performance-engineer: Drills down into the identified service, profiling code and identifying specific bottlenecks.
4. @database-optimizer: If the issue is query-related, this agent analyzes execution plans and recommends indexing changes.
5. @python-pro (or other language specialist): Implements the code-level fix based on the findings.
6. @incident-responder: Manages the deployment of the fix and confirms the resolution before closing the incident.

Contributing

Contributions are welcomed! If you wish to add or improve an agent, please ensure it aligns with the core philosophy of principal-level expertise. Focus on strategic value, clear deliverables, and modern best practices.

License

This collection of agents is licensed under the MIT License.