CrewAI YouTube Video to Blog

This project leverages the CrewAI framework to create a blog post from YouTube video content. It uses AI agents to research video transcriptions on a specified topic from a YouTube channel and generate a summarized blog post. The application focuses on topics like AI, Machine Learning, Deep Learning, and Data Science, using the @krishnaik06 YouTube channel as the primary source.

Project Overview

• Purpose: Automate blog content creation by extracting and summarizing information from YouTube videos.
• Framework: CrewAI, utilizing AI agents for research and writing tasks.
• Key Features:
  • Research agent fetches video transcriptions for a given topic.
  • Writing agent creates a blog post based on the research.
  • Output is saved as a markdown file (new-blog-post.md).
• GitHub Repository: CrewAI_YtVideo_to_Blog

Topics

• AI content generation
• YouTube video transcription
• CrewAI framework
• Blog automation
• Data Science and Machine Learning

Tech Stack

• Language: Python
• Framework: CrewAI
• Libraries: crewai, crewai_tools, python-dotenv
• APIs: OpenAI (GPT-4), YouTube Channel Search Tool
• Tools: GitHub

File Structure

CrewAI_YtVideo_to_Blog/
├── .gitignore          # Ignored files for Git
├── LICENSE            # Project license
├── README.md          # Project documentation
├── agents.py          # Defines AI agents for research and writing
├── crew.py            # Configures CrewAI and task execution
├── requirements.txt   # Python dependencies
├── tasks.py           # Defines research and writing tasks
├── tools.py           # Configures YouTube search tool

Prerequisites

• Python 3.8+
• OpenAI API key (set in .env file)
• GitHub repository cloned: CrewAI_YtVideo_to_Blog
• YouTube channel handle (@krishnaik06) for video content

Installation

1. Clone the Repository:

   git clone https://github.com/Monish-Nallagondalla/CrewAI_YtVideo_to_Blog.git
   cd CrewAI_YtVideo_to_Blog

2. Install Dependencies:

   pip install -r requirements.txt

3. Set Up Environment Variables:

   • Create a .env file in the project root.
   • Add your OpenAI API key:

     OPENAI_API_KEY=your_openai_api_key

Usage

1. Configure Agents and Tasks:

   • agents.py: Defines a research agent to fetch YouTube video transcriptions and a writing agent to create blog content.
   • tasks.py: Specifies tasks for researching video content and writing a summarized blog post.
   • tools.py: Configures the YouTube search tool for the @krishnaik06 channel.
   • crew.py: Sets up the CrewAI framework with sequential task execution.

2. Run the Application:

   python crew.py

   • The script processes the topic AI VS ML VS DL vs Data Science by default.
   • Output is saved as new-blog-post.md in the project directory.

3. Customize Topic:

   • Modify the topic in crew.py by updating the inputs dictionary:

     result = crew.kickoff(inputs={'topic': 'Your_New_Topic'})

Code Details

• agents.py:
  • blog_researcher: Fetches video transcriptions for the specified topic using the YouTube search tool.
  • blog_writer: Generates a blog post summarizing the video content.
• tasks.py:
  • research_task: Produces a 3-paragraph report based on video content.
  • write_task: Creates a summarized blog post saved as new-blog-post.md.
• tools.py:
  • Configures YoutubeChannelSearchTool for the @krishnaik06 channel.
• crew.py:
  • Orchestrates agents and tasks using CrewAI's sequential process.
  • Configures memory, caching, and a maximum RPM of 100.

Output

• The generated blog post is saved as new-blog-post.md in the project directory.
• The research task produces a 3-paragraph report, and the writing task summarizes it into a blog post.

Dependencies

Listed in requirements.txt:

• crewai
• crewai_tools
• python-dotenv

Notes

• Ensure a valid OpenAI API key is set in the .env file.
• The YouTube search tool requires internet access to fetch video transcriptions.
• The application is configured for the @krishnaik06 YouTube channel; modify tools.py to use a different channel.
• The OpenAI model used is gpt-4-0125-preview; update agents.py if a different model is preferred.

Contributing

1. Fork the repository.
2. Create a feature branch (git checkout -b feature-branch).
3. Commit changes (git commit -m "Add feature").
4. Push to the branch (git push origin feature-branch).
5. Open a pull request.

License

This project is licensed under the MIT License. See the LICENSE file for details.

Contact

For questions or issues, open a GitHub issue or contact the repository owner at Monish-Nallagondalla.