crewai-tools 0.0.3

Set of tools for the crewAI framework

crewAI Tools

This document provides a comprehensive guide for setting up sophisticated tools for crewAI agents, facilitating the creation of bespoke tooling to empower your AI solutions.

In the realm of CrewAI agents, tools are pivotal for enhancing functionality. This guide outlines the steps to equip your agents with an arsenal of ready-to-use tools and the methodology to craft your own.

Homepage | Documentation | Chat with Docs | Examples | Discord

Table of contents

• Creating Your Tools
  • Subclassing BaseTool
  • Functional Tool Creation
  • Utilizing the tool Decorator
• Contribution Guidelines
• Development Setup

Creating Your Tools

Tools are always expect to return strings, as they are meant to be used by the agents to generate responses.

There are three ways to create tools for crewAI agents:

• Subclassing BaseTool
• Creating a tool from a function or lambda
• Using the tool decorator

Subclassing BaseTool

class MyCustomTool(BaseTool):
    name: str = "Name of my tool"
    description: str = "Clear description for what this tool is useful for, you agent will need this information to use it."

    def _run(self, argument) -> str:
        # Implementation goes here
        pass

Define a new class inheriting from BaseTool, specifying name, description, and the _run method for operational logic.

Functional Tool Creation

my_tool = Tool(
    name="Name of my tool"
    description="Clear description for what this tool is useful for, you agent will need this information to use it.",
    func=lambda argument: # Your function logic here
)

For a simpler approach, create a Tool object directly with the required attributes and a functional logic.

Utilizing the tool Decorator

@tool("Name of my tool")
def my_tool(question: str) -> str:
    """Clear description for what this tool is useful for, you agent will need this information to use it."""
    # Function logic here

The tool decorator simplifies the process, transforming functions into tools with minimal overhead.

Contribution Guidelines

We eagerly welcome contributions to enrich this toolset. To contribute:

1. Fork the Repository: Begin with forking the repository to your GitHub account.
2. Feature Branch: Create a new branch in your fork for the feature or improvement.
3. Implement Your Feature: Add your contribution to the new branch.
4. Pull Request: Submit a pull request from your feature branch to the main repository.

Your contributions are greatly appreciated and will help enhance this project.

Development Setup

Installing Dependencies:

poetry install

Activating Virtual Environment:

poetry shell

Setting Up Pre-commit Hooks:

pre-commit install

Running Tests:

poetry run pytest

Static Type Checking:

poetry run pyright

Packaging:

poetry build

Local Installation:

pip install dist/*.tar.gz

Thank you for your interest in enhancing the capabilities of AI agents through advanced tooling. Your contributions make a significant impact.