Crewai Claude

🚀 Crewai Team

Welcome to the Crewai Team project, powered by crewAI. This template is designed to help you easily set up a multi - agent AI system, leveraging the powerful and flexible framework of crewAI. Our goal is to enable your agents to collaborate effectively, complete complex tasks, and maximize their collective wisdom and capabilities.

🚀 Quick Start

📦 Installation

Ensure that Python >=3.10 <3.13 is installed on your system. This project uses UV for dependency management, providing a seamless setup and execution experience.

First, install uv if you haven't already:

pip install uv

Next, install the dependencies in the project directory: (Optional) Lock the dependencies and install them using the CLI command:

crewai install

Customization

Add your OPENAI_API_KEY to the .env file

• Modify src/crewai/config/agents.yaml to define your agents.
• Modify src/crewai/config/tasks.yaml to define your tasks.
• Modify src/crewai/crew.py to add your own logic, tools, and specific parameters.
• Modify src/crewai/main.py to add custom inputs for agents and tasks.

💻 Usage Examples

Running the Project

Sequential Team

To start the Crewai team and begin task execution, run the following from the project's root folder:

$ crewai run

This command initializes the Crewai Crew, assembles the agents, and assigns tasks as you've defined in the configuration.

The unmodified example will create a report.md file containing the research results from the LLM in the root directory.

Hierarchical Team

The project also includes a hierarchical implementation where each agent specializes in using specific tools. To run the hierarchical team:

$ hierarchical

Or:

$ run_hierarchical

This will create a hierarchical_result.md file showing the output from the hierarchical process.

For more information about the hierarchical implementation, refer to src/crewai/hierarchical/README.md.

Model Control Protocol (MCP) Integration

The project includes an MCP server that exposes CrewAI tools via a REST API. This allows Claude and other LLMs to access and utilize CrewAI tools.

Starting the MCP Server

$ start_mcp

Or run it directly:

$ python -m mcp.run_server

By default, the server runs on 0.0.0.0:8000. You can customize it as follows:

$ start_mcp --host 127.0.0.1 --port 9000

Available MCP Tools

The MCP server provides access to the following tools:

• Custom CrewAI tools
• Web search functionality
• Data analysis capabilities

For more details, refer to mcp/README.md.

📚 Documentation

Understanding Your Team

The Crewai team consists of multiple AI agents, each with a unique role, goal, and set of tools. These agents collaborate on tasks defined in config/tasks.yaml, leveraging their collective skills to achieve complex goals. The config/agents.yaml file outlines the capabilities and configurations of each agent in your team.

Support

For questions or feedback regarding the Crewai team or crewAI, visit our documentation, GitHub repository, or join our Discord server and chat with our documentation.

Let's create wonders together with the power and simplicity of crewAI.