This repository provides a comprehensive cookbook of code templates and guides for utilizing CrewAI's agentic workflow implementations. It is targeted at developers looking to build sophisticated AI-driven applications, offering easily integrable code snippets and practical examples for various use cases.

How It Works

The quickstart showcases CrewAI's capabilities through a variety of notebook examples, demonstrating both sequential and hierarchical agentic workflows. It highlights the integration of numerous tools, including RAG-based search functionalities for diverse data formats (TXT, CSV, PDF, JSON, XML, MDX, DOCX), code interpretation, web scraping, and specialized API interactions. This modular approach allows users to quickly adopt and adapt specific functionalities for their projects.

Quick Start & Requirements

• Prerequisites: Familiarity with CrewAI (docs: https://docs.crewai.com/) and an API key from LLM providers (Anthropic, OpenAI, Groq, Cohere).
• Examples: Notebooks are available for sequential and hierarchical workflows, tool usage (e.g., TXTSearchTool, CodeDocsSearchTool, CodeInterpreterTool, GithubSearchTool), custom tools, Python scripts, and Streamlit GUIs. Local LLM examples with Ollama are also provided.

Highlighted Details

• Extensive collection of tools for RAG, code interpretation, web scraping, and more.
• Demonstrates both sequential and hierarchical agentic workflow patterns.
• Includes examples for building GUIs with Streamlit and integrating local LLMs.
• Provides Python scripts for direct integration.

Maintenance & Community

The project encourages community contributions for new examples and improvements. Users can share ideas and report issues via the GitHub issues page.

Licensing & Compatibility

• License: MIT License.
• Compatibility: Permissive MIT license allows for commercial use and integration into closed-source projects.

Limitations & Caveats

The quickstart assumes a foundational understanding of CrewAI and requires API keys for LLM providers, which may incur costs. Some examples might require specific environment setups or data files not included in the repository.