Task Manager RBAC - Prompt Engineering Repository

This repository contains comprehensive prompt engineering documentation for building a serverless task management system with role-based access control using Amazon Q CLI and AWS services.

🚀 Live Implementation: See the actual working application at task-manager-role-based-access

📋 What's Inside

This repository demonstrates how to use structured prompts and formal specifications to generate enterprise-grade applications with AI assistance.

📄 Documentation Files

• DETAILED_PROJECT_PROMPT.md - Complete project specification with RFC 2119 compliant requirements for AI input
• RFC2119_REQUIREMENTS.md - Formal requirements document using RFC 2119 terminology (MUST, SHOULD, MAY)

🎯 Prompt Engineering Approach

This project demonstrates how to:

1. Structured Requirements

• Use RFC 2119 terminology for precise AI instructions
• Define clear role-based permissions (Admin, Contributor, Viewer)
• Specify exact technical requirements for each component

2. Architecture-First Prompting

• Break complex systems into manageable components
• Define AWS service requirements upfront
• Ensure security and scalability from the start

3. Role-Based System Design

• Admin: Full CRUD operations, user management, system configuration
• Contributor: Own task management, assigned task access, file operations
• Viewer: Read-only access to assigned tasks only

🏗️ Generated System Architecture

The prompts in this repository generate:

• Frontend: React SPA with TypeScript, AWS Amplify
• Backend: AWS Lambda, API Gateway, DynamoDB
• Infrastructure: AWS CDK (TypeScript)
• Authentication: AWS Cognito with JWT tokens
• Storage: Amazon S3 for file attachments
• Security: Custom Lambda authorizers, encryption, IAM policies

🔧 How to Use These Prompts

1. Start with the detailed prompt - Use DETAILED_PROJECT_PROMPT.md as input to Amazon Q CLI
2. Reference RFC 2119 requirements - Use formal specifications for precise AI guidance
3. Generate incrementally - Build system components step by step
4. Validate outputs - Ensure generated code meets all MUST requirements

Example Usage with Amazon Q CLI

# Use the detailed prompt as input
q generate --input DETAILED_PROJECT_PROMPT.md --type infrastructure

# Generate specific components
q generate --input "Create Lambda authorizer based on User Role System section"

# Validate against requirements
q validate --requirements RFC2119_REQUIREMENTS.md

📊 Prompt Engineering Results

Using these structured prompts with Amazon Q CLI generates:

• ✅ Complete AWS CDK Infrastructure (200+ lines)
• ✅ Production-ready Lambda Functions with proper error handling
• ✅ Type-safe React Components with role-based rendering
• ✅ Comprehensive Security Implementation following AWS best practices
• ✅ Formal Documentation and API specifications

🎓 Key Learnings

Effective Prompt Patterns

1. Constraint-Based Prompting - Define what roles MUST NOT do
2. Security-First Approach - Lead with security requirements
3. Formal Specifications - Use RFC 2119 for precision
4. Incremental Building - Layer complex systems through connected prompts

Best Practices

• Start with comprehensive requirements documents
• Use formal specification languages (RFC 2119, OpenAPI)
• Define role boundaries explicitly
• Validate AI outputs against requirements

🔗 Related Resources

• Working Implementation - See the generated code in action
• AWS CDK Documentation - Infrastructure as Code reference
• RFC 2119 - Requirement levels specification
• Amazon Q CLI Documentation - AI-assisted development guide

🤝 Contributing

Contributions to improve the prompt engineering approach are welcome:

1. Review the existing prompt structures
2. Test prompts with Amazon Q CLI
3. Suggest improvements to requirement specifications
4. Share results and learnings

📄 License

This project is licensed under the MIT License.

---

Note: This repository focuses on prompt engineering and AI-assisted development. For the actual implementation and deployable code, visit the main project repository.