Learning CrewAI Repository

Welcome to the comprehensive Learning CrewAI Repository! This collection showcases three progressively advanced CrewAI projects that demonstrate the full spectrum of AI agent capabilities, from basic research workflows to production-ready business intelligence systems.

📋 Repository Overview

This repository contains 3 specialized CrewAI crews that demonstrate different aspects of multi-agent AI systems:

| Crew | Complexity | Agents | Tools | Use Case | |------|------------|--------|-------|----------| | 01_learning_crewai_project | Basic | 2 | 0 | Research & Reporting | | 02_ai_news | Advanced | 3 | 2 | News Intelligence | | 03_dfw_business_leads | Production | 5 | 6 | Business Lead Generation |

🤖 Crews Overview

1️⃣ Learning CrewAI Project

📁 Directory: 01_learning_crewai_project/
🎯 Purpose: Introduction to CrewAI fundamentals
👥 Agents: 2 (Researcher, Analyst)
🔧 Tools: None (base LLM capabilities)
📊 Output: Research reports on any topic

Perfect for:

• First-time CrewAI users
• Understanding agent collaboration
• Learning task dependencies
• Basic report generation

📖 Full Documentation →

---

2️⃣ AI News Intelligence Crew

📁 Directory: 02_ai_news/
🎯 Purpose: Advanced AI and developer tools news intelligence
👥 Agents: 3 (Research Specialist, Web Search Specialist, Reporter)
🔧 Tools: 2 (ScrapeWebsiteTool, BraveSearchTool)
📊 Output: Comprehensive AI news reports with source citations

Perfect for:

• Technology professionals tracking AI trends
• Learning advanced tool integration
• Multi-agent collaboration patterns
• Real-time data gathering

📖 Full Documentation →

---

3️⃣ North Fort Worth Business Leads Discovery Crew

📁 Directory: 03_dfw_business_leads/
🎯 Purpose: Production-ready business intelligence and lead generation
👥 Agents: 5 (Scout, Analyzer, Qualifier, Researcher, Writer)
🔧 Tools: 6 (Business search, web analysis, lead qualification)
📊 Output: Actionable business leads reports with revenue projections

Perfect for:

• Web development agencies and freelancers
• Sales and business development teams
• Complex multi-agent workflows
• Production business applications

📖 Full Documentation →

🚀 Quick Start Guide

Prerequisites

• Python: >=3.10 <=3.13
• UV: Modern Python package manager
• Ollama: (Recommended) For local LLM execution

Installation Steps

1. Install UV (if not already installed):

   pip install uv
   

2. Clone and setup:

   git clone <repository-url>
   cd Learning-CrewAI
   

3. Choose your crew and navigate to its directory:

   # For beginners
   cd 01_learning_crewai_project
   
   # For advanced users
   cd 02_ai_news
   
   # For production use
   cd 03_dfw_business_leads
   

4. Install dependencies:

   uv sync
   

5. Configure environment (create .env file):

   # Basic configuration for local LLM
   MODEL=ollama/mistral:latest
   OPENAI_API_BASE=http://localhost:11434
   OPENAI_API_KEY=ollama
   OPENAI_MODEL_NAME=ollama/mistral:latest
   
   # For crews requiring Brave Search
   BRAVE_API_KEY=your_brave_api_key_here
   

6. Run the crew:

   crewai run
   

🏗️ Repository Structure

Learning-CrewAI/
├── README.md                          # This file
├── 01_learning_crewai_project/        # Basic crew
│   ├── README.md                      # Detailed documentation
│   ├── src/learning_crewai_project/
│   │   ├── config/
│   │   │   ├── agents.yaml           # Agent definitions
│   │   │   └── tasks.yaml            # Task definitions  
│   │   ├── crew.py                   # Crew orchestration
│   │   ├── main.py                   # Entry point
│   │   └── tools/                    # Custom tools (basic)
│   └── report.md                     # Generated output
├── 02_ai_news/                       # Advanced crew
│   ├── README.md                     # Detailed documentation
│   ├── src/ai_news/
│   │   ├── config/
│   │   │   ├── agents.yaml           # Agent definitions
│   │   │   └── tasks.yaml            # Task definitions
│   │   ├── crew.py                   # Crew orchestration
│   │   ├── main.py                   # Entry point
│   │   └── tools/                    # Custom tools
│   └── Reports/                      # Generated reports
│       └── AI_News_Report_*.md
└── 03_dfw_business_leads/            # Production crew
    ├── README.md                     # Detailed documentation
    ├── src/dfw_business_leads/
    │   ├── config/
    │   │   ├── agents.yaml          # Agent definitions
    │   │   └── tasks.yaml           # Task definitions
    │   ├── crew.py                  # Crew orchestration
    │   ├── main.py                  # Entry point
    │   └── tools/                   # Advanced business tools
    │       ├── business_search_tools.py
    │       └── brave_search_tool.py
    ├── outputs/                     # Business reports
    │   └── north_fort_worth_leads_report.md
    └── test_crew.py                 # Comprehensive testing

🛠️ Local LLM Setup (Recommended)

Using Ollama

1. Install Ollama:

   # macOS
   brew install ollama
   
   # Linux
   curl -fsSL https://ollama.ai/install.sh | sh
   

2. Start Ollama service:

   ollama serve
   

3. Pull recommended models:

   # Lightweight, fast model
   ollama pull mistral:latest
   
   # More powerful model  
   ollama pull llama3.2:latest
   
   # Coding-focused model
   ollama pull deepseek-coder-v2:latest
   

4. Configure crews to use local models (already configured in examples)

Benefits of Local LLMs

• 🔒 Privacy: Your data never leaves your machine
• 💰 Cost-effective: No API fees or usage limits
• ⚡ Speed: No network latency for requests
• 🎛️ Control: Full control over model behavior
• 🔄 Reliability: No API downtime or rate limits

🔧 Advanced Configuration

Environment Variables

Each crew supports different configuration options:

# LLM Configuration
MODEL=ollama/mistral:latest                    # Model identifier
OPENAI_API_BASE=http://localhost:11434         # Ollama endpoint
OPENAI_API_KEY=ollama                          # Dummy key for Ollama
OPENAI_MODEL_NAME=ollama/mistral:latest        # Full model name

# External APIs (when needed)
BRAVE_API_KEY=your_brave_api_key_here          # For web search
SERPER_API_KEY=your_serper_key_here            # Alternative search
OPENAI_API_KEY=your_openai_key_here            # For cloud LLMs

# Performance Tuning
CREW_VERBOSE=true                              # Detailed logging
CREW_MEMORY=true                               # Enable agent memory
CREW_PLANNING=true                             # Enable planning feature

Model Recommendations

| Use Case | Recommended Model | Size | Speed | Quality | |----------|------------------|------|-------|---------| | Development & Testing | mistral:latest | 4.1GB | Fast | Good | | Production Quality | llama3.2:latest | 2.0GB | Medium | Excellent | | Code Generation | deepseek-coder-v2:latest | 8.9GB | Medium | Excellent | | Reasoning Tasks | phi4-reasoning:latest | 11GB | Slow | Excellent |

🎓 Learning Path

Beginner Path

1. Start with: 01_learning_crewai_project

   • Learn basic agent configuration
   • Understand task dependencies
   • Practice with simple workflows

2. Learn concepts:

   • Agent roles, goals, and backstories
   • Task descriptions and expected outputs
   • Sequential process execution

Intermediate Path

1. Move to: 02_ai_news

   • Integrate external tools
   • Handle API configuration
   • Work with real-time data

2. Learn concepts:

   • Tool integration patterns
   • Multi-agent collaboration
   • Error handling and reliability

Advanced Path

1. Master: 03_dfw_business_leads

   • Complex business logic
   • Production-ready systems
   • Custom tool development

2. Learn concepts:

   • Business intelligence workflows
   • Lead qualification systems
   • Market research automation

Expert Level

1. Extend and customize:
   • Adapt crews for your use cases
   • Integrate with business systems
   • Scale to production environments

🔄 Common Workflows

Research & Analysis

# Basic research (01_learning_crewai_project)
cd 01_learning_crewai_project
crewai run  # Researches AI LLMs

# Advanced news intelligence (02_ai_news)  
cd ../02_ai_news
crewai run  # Generates AI news report

Business Intelligence

# Lead generation (03_dfw_business_leads)
cd ../03_dfw_business_leads
python test_crew.py  # Test functionality
crewai run           # Generate business leads

Custom Development

# Create new crew based on examples
crewai create crew my-custom-crew
# Copy patterns from existing crews
# Adapt agents, tasks, and tools

🧪 Testing & Development

Running Tests

Each crew includes testing capabilities:

# Basic functionality tests
cd 01_learning_crewai_project && crewai run

# Advanced tool integration tests  
cd 02_ai_news && crewai run

# Comprehensive business logic tests
cd 03_dfw_business_leads && python test_crew.py

Development Tips

1. Start Simple: Begin with basic crews and add complexity
2. Test Incrementally: Verify each component before integration
3. Use Local LLMs: Faster iteration and no API costs
4. Monitor Performance: Watch agent interactions and tool usage
5. Handle Errors: Implement proper error handling and fallbacks

📚 Key CrewAI Concepts Demonstrated

Agent Design Patterns

• Specialist Agents: Each agent has a specific expertise area
• Tool Integration: Agents equipped with appropriate tools
• Collaboration: Agents work together through task dependencies

Task Orchestration

• Sequential Processing: Tasks build on previous results
• Parallel Execution: Independent tasks run simultaneously
• Output Management: Structured outputs and file generation

Tool Architecture

• Custom Tools: Business-specific functionality
• API Integration: External service integration
• Error Handling: Graceful failure and recovery

Production Considerations

• Configuration Management: Environment-based configuration
• Performance Optimization: Efficient tool usage and caching
• Scalability: Patterns for scaling to larger systems

🌟 Use Cases & Applications

Business Applications

• Lead Generation: Automated prospect discovery and qualification
• Market Research: Competitive analysis and trend monitoring
• Content Creation: Research-backed content generation
• Customer Intelligence: Customer research and segmentation

Technical Applications

• News Monitoring: Technology trend tracking
• Code Analysis: Automated code review and documentation
• System Integration: Multi-system data aggregation
• Process Automation: Complex workflow automation

Research Applications

• Academic Research: Literature review and synthesis
• Market Analysis: Industry research and reporting
• Competitive Intelligence: Competitor monitoring
• Trend Analysis: Pattern recognition and forecasting

🤝 Contributing

We welcome contributions! Here's how to get involved:

1. Fork the repository
2. Create a new branch: git checkout -b feature/your-feature
3. Make your changes and add tests
4. Commit your changes: git commit -m 'Add your feature'
5. Push to the branch: git push origin feature/your-feature
6. Submit a pull request

Contribution Ideas

• New crew examples for different industries
• Additional tools and integrations
• Performance optimizations
• Documentation improvements
• Testing enhancements

📖 Additional Resources

CrewAI Documentation

• Official Docs: https://docs.crewai.com
• GitHub Repository: https://github.com/joaomdmoura/crewai
• Discord Community: Join Discord
• Interactive Chat: Chat with docs

Local LLM Resources

• Ollama: https://ollama.ai
• Model Library: https://ollama.ai/library
• GPU Acceleration: CUDA/ROCm setup guides

External APIs

• Brave Search: https://brave.com/search/api/
• OpenAI API: https://platform.openai.com
• Anthropic API: https://console.anthropic.com

🆘 Support & Community

Getting Help

• Documentation Issues: Check individual crew README files
• Technical Problems: Review configuration and error messages
• Feature Requests: Submit GitHub issues with detailed descriptions
• General Questions: Join the CrewAI Discord community

Troubleshooting Common Issues

LLM Connection Problems:

# Check Ollama status
ollama ps

# Restart Ollama service  
ollama serve

# Verify model availability
ollama list

Tool Integration Errors:

• Verify API keys in .env files
• Check network connectivity
• Review tool documentation for requirements

Performance Issues:

• Use appropriate models for your hardware
• Monitor system resources during execution
• Consider batch processing for large datasets

---

🎯 Next Steps

1. Choose your starting point based on experience level
2. Follow the learning path progression
3. Experiment with modifications and customizations
4. Build your own crews using these patterns
5. Share your creations with the community

Ready to explore the future of AI agent collaboration? Pick a crew and start building! 🚀🤖

---

Happy building with CrewAI! 🎉