AI Content Agent - Agentube

Live Link

https://agentube-edasgh.vercel.app/

Overview

Agentube is an AI-powered platform that helps content creators analyze, transcribe, and optimize their YouTube videos. The tool leverages artificial intelligence to provide deep insights, auto-generate video scripts, create engaging thumbnails, and more.

Features

• AI Analysis: Gain insights into viewer engagement and content quality.
• Auto Transcription: Convert YouTube videos into accurate transcripts.
• AI Script Generator: Generate engaging scripts for YouTube videos. (Coming Soon!)
• Thumbnail Creator: Automatically create high-quality thumbnails.
• Engagement Metrics: Track views, likes, comments, and retention.

How It Works

1. Connect with your content: Share your YouTube video URL.
2. Get AI-Powered Insights: Extract transcripts, analyze engagement, and generate a script.
3. Receive Ready-to-Use Content: Download the transcript, script, and AI-generated thumbnail.

Tech Stack

• Frontend: Next.js, TypeScript, Tailwind CSS
• Backend: Node.js, Express.js
• Database: Convex
• Authentication: Clerk
• State Management: Redux Toolkit
• AI Integration: Gemini, Together AI, Vercel AI SDK

Installation

1. Clone the repository:

   git clone https://github.com/Edasgh/agentube.git
   cd ai-content-agent

2. Install dependencies:

   npm install

3. Create a .env.local file and configure environment variables:

        # Clerk
   
        NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY=pk_test_your_clerk_publishable_key
        CLERK_SECRET_KEY=sk_test_your_clerk_secret_key
   
        # Schematic
   
        NEXT_PUBLIC_SCHEMATIC_PUBLISHABLE_KEY=api_your_schematic_publishable_key
        SCHEMATIC_API_KEY=sch_dev_your_schematic_api_key
   
        # Youtube API
   
        YOUTUBE_API_KEY=your_youtube_api_key
   
        # Deployment used by `npx convex dev`
   
        CONVEX_DEPLOYMENT=dev:your-convex-deployment # team: your-team, project: your-project
   
        NEXT_PUBLIC_CONVEX_URL=https://your-convex-deployment.convex.cloud
   
        # Gemini
   
        GOOGLE_GENERATIVE_AI_API_KEY=your_gemini_api_key
   
        #together ai api key
        TOGETHER_API_KEY=sk-ant-api03-your_together_ai_api_key
   
        # BASE URL
   
        NEXT_PUBLIC_BASE_URL="http://localhost:3000"

4. Run the development server:
```sh
npm run dev

5. Open http://localhost:3000 in your browser.

Deployment

You can deploy the project using Vercel:

vercel

Ensure you have set up the required environment variables in Vercel.

Contributing

Feel free to contribute by submitting issues or pull requests. Make sure to follow the project's coding standards.

License

This project is licensed under the MIT License.

Contact

For questions or support, reach out via My Email.