CrewAI Series

This repository contains a series of projects demonstrating the capabilities and evolution of CrewAI, a framework for orchestrating role-playing AI agents. Each day's project builds upon the previous, showcasing different aspects and features of CrewAI.

Overview of Daily Projects

Day 01: Introduction to CrewAI

• Basic setup of CrewAI
• Creating simple agents and tasks
• Understanding the fundamental concepts of CrewAI

Day 02: Multi-Agent Collaboration

• Implementing multiple agents with distinct roles
• Designing tasks that require agent collaboration
• Exploring basic inter-agent communication

Day 03: Advanced Task Management

• Introducing more complex task structures
• Implementing task dependencies and sequences
• Enhancing agent decision-making capabilities

Day 04: Automated News Analysis

• Creating a system for news gathering and analysis
• Implementing agents for research, analysis, and report writing
• Generating comprehensive news reports automatically

Day 05: Content Marketing Strategy

• Developing an automated content marketing workflow
• Creating specialized agents for content strategy, writing, editing, and SEO
• Producing a complete content marketing strategy with AI assistance

Day 06: CrewAI Flow System

• Introduction to the new Flow system in CrewAI
• Implementing multiple crews with different configurations
• Using YAML files for agent and task definitions
• Visualizing workflows with the flow.plot() method

Getting Started

Each day's project is contained in its own directory (e.g., day_01, day_02, etc.). To get started with a specific day:

1. Navigate to the day's directory
2. Follow the README instructions in that directory for setup and execution

Prerequisites

• Python >=3.10 <=3.13
• Poetry (for dependency management)

General Setup

1. Clone this repository
2. Install Poetry if you haven't already:

   pip install poetry
   

3. Navigate to the specific day's directory
4. Install dependencies:

   poetry install
   

5. Set up your OpenAI API key in the .env file

Running a Project

Typically, you can run a day's project with:

poetry run python src/day_XX/main.py

Replace XX with the day number (e.g., 01, 02, etc.)

Jupyter Notebooks

This series also includes Jupyter notebooks for interactive exploration and execution of CrewAI concepts. Here's a list of the available notebooks:

1. day_01.ipynb: Introduction to CrewAI

   • This notebook demonstrates the basic setup and usage of CrewAI, introducing simple agents and tasks.

2. day_02.ipynb: Local LLM Agents

   • shows how to use Ollama for creating agents and tasks

3. day_03.ipynb: News Aggregator start

   • Starts the news aggregator project with Tools

4. day_04.ipynb: Custom Tooling

   • Shows how to create custom tools for the agents using a more refined SerperDev tool

5. day_05.ipynb: New Feature in CrewAI

   • Introduces the new Flow system in CrewAI, implementing multiple crews and using YAML files for configurations.

Running the Notebooks

To run the Jupyter notebooks:

1. Ensure you have Jupyter installed. If not, you can install it using:

   pip install jupyter
   

2. Navigate to the project directory and start Jupyter:

   jupyter notebook
   

3. In the Jupyter interface, navigate to the desired notebook (e.g., day_01.ipynb) and open it.

4. You can run the cells individually by clicking on them and pressing Shift+Enter, or you can run all cells from the "Cell" menu by selecting "Run All".

5. Make sure you have set up your environment variables (like OPENAI_API_KEY) before running the notebooks.

Each notebook corresponds to its respective day's project, allowing for an interactive exploration of the concepts covered in that day's lesson.