Spaces:
Running
Running
title: Course Creator AI - Multi-Tool Agent Demo | |
emoji: ๐ | |
colorFrom: blue | |
colorTo: purple | |
sdk: gradio | |
sdk_version: 5.33.1 | |
app_file: app.py | |
pinned: false | |
license: mit | |
tags: | |
- agent-demo-track | |
- education | |
- ai-agents | |
- course-generation | |
- multi-tool | |
- web-research | |
- image-generation | |
- gradio | |
- hackathon | |
short_description: AI-powered educational course generator | |
# ๐ Course Creator AI - Multi-Tool Agent Demo | |
**Track 3: Agentic Demo Submission** | **Hugging Face Agent Hackathon 2025** | |
> **๐ฏ Tag**: `agent-demo-track` | |
## ๐ฌ Video Overview | |
**๐น [Watch Demo Video - App Usage & Purpose Explanation](https://youtu.be/3VS010g4_c8)** | |
--- | |
## ๐ What is Course Creator AI? | |
Course Creator AI is a **real multi-tool agent system** that demonstrates the power of AI agents working together to create comprehensive educational content. This isn't a simulation - it's a live demonstration of agents using actual tools and APIs to research, generate, and create educational materials. | |
### ๐ฏ **Purpose & Vision** | |
Transform any topic into a complete, engaging educational experience through intelligent agent orchestration. Perfect for educators, students, content creators, and anyone looking to learn or teach effectively. | |
## ๐ค **Real Agent Workflow in Action** | |
Watch our agents work together in real-time: | |
### **๐ Research Agent** | |
- **Tool**: DuckDuckGo Search API | |
- **Action**: Searches the web for current, relevant information | |
- **Output**: Curated research data and reliable sources | |
### **๐ Content Agent** | |
- **Tool**: Advanced LLM Integration (OpenAI, Anthropic, Google Gemini) | |
- **Action**: Processes research into structured educational content | |
- **Output**: Comprehensive lessons with learning objectives, interactive flashcards, and quiz | |
### **๐จ Image Agent** | |
- **Tool**: Pollinations API (GPT-Image Model) | |
- **Action**: Generates contextual educational images | |
- **Output**: High-quality, relevant visual aids | |
### **๐ง Assessment Agent** | |
- **Tool**: Intelligent Content Analysis | |
- **Action**: Creates interactive learning materials | |
- **Output**: Flashcards and quizzes for knowledge retention | |
## โจ **Key Features** | |
### ๐ **Complete Course Generation** | |
- **5 Comprehensive Lessons** with detailed content (2000+ words each) | |
- **Learning Objectives** tailored to difficulty level | |
- **Real-world Examples** and practical applications | |
- **Progressive Difficulty** building from fundamentals | |
### ๐ **Interactive Learning Materials** | |
- **15 Smart Flashcards** for key concept reinforcement | |
- **10 Multiple-Choice Questions** with detailed explanations | |
- **Immediate Feedback** and scoring system | |
- **Category-based Organization** for focused study | |
### ๐ผ๏ธ **Visual Learning Enhancement** | |
- **AI-Generated Educational Images** using Flux model | |
- **Contextual Visual Aids** that enhance understanding | |
- **No Watermarks** - clean, professional imagery | |
- **Smart Placeholder System** for optimal image placement | |
### ๐ง **Multi-Provider LLM Support** | |
- **OpenAI GPT Models** for premium quality | |
- **Anthropic Claude** for detailed analysis | |
- **Google Gemini** for cost-effective generation | |
- **OpenAI-Compatible APIs** for flexibility | |
- **Smart Provider Switching** via environment variables | |
## ๐ ๏ธ **Technical Architecture** | |
### **Agent Orchestration** | |
``` | |
Research Agent โ Content Agent โ Assessment Agent โ Image Agent | |
โ โ โ โ | |
Web Search โ Lesson Creation โ Quiz Generation โ Visual Enhancement | |
``` | |
### **Real Tool Integration** | |
- **๐ Web Research**: Live DuckDuckGo search with Crawl4AI content extraction | |
- **๐ง LLM Processing**: Multi-provider support with smart JSON parsing | |
- **๐จ Image Generation**: Pollinations API with Flux model | |
- **๐ Content Analysis**: Intelligent educational content structuring | |
### **Smart Features** | |
- **Markdown-Aware JSON Parser** handles LLM response variations | |
- **Provider Fallback System** ensures reliability | |
- **Real-time Progress Tracking** (console-based) | |
- **Error Recovery** with graceful degradation | |
## ๐ฎ **How to Use** | |
### **1. Quick Start** | |
1. **Enter Your Topic**: Anything from "Python Programming" to "Ancient History" | |
2. **Choose Difficulty**: Beginner, Intermediate, or Advanced | |
3. **Set Lesson Count**: 3-7 lessons based on your needs | |
4. **Click Generate**: Watch the agents work their magic! | |
### **2. Explore Your Course** | |
- **๐ Course Overview**: See the complete structure and learning objectives | |
- **๐ Lessons**: Dive into detailed, research-backed content | |
- **๐ Flashcards**: Test your knowledge with interactive cards | |
- **๐ Quiz**: Challenge yourself with comprehensive questions | |
- **๐จ Images**: Enjoy AI-generated visual learning aids | |
### **3. Interactive Learning** | |
- **Flip Flashcards** to reveal answers | |
- **Take the Quiz** with immediate feedback | |
- **Review Explanations** for deeper understanding | |
- **Track Your Progress** through the material | |
## ๐ง **Configuration** | |
### **๐ Quick Start (Hugging Face Space)** | |
1. **Select LLM Provider**: Choose from OpenAI, Anthropic, Google Gemini, or OpenAI-Compatible | |
2. **Enter API Key**: Add your API key in the interface | |
3. **Generate Course**: Enter any topic and start creating! | |
### **Provider Options** | |
- **๐ Google Gemini**: Free tier available, excellent quality | |
- **๐ OpenAI GPT**: Premium quality, requires paid API | |
- **๐ง Anthropic Claude**: Advanced reasoning, requires paid API | |
- **๐ OpenAI-Compatible**: Use any compatible endpoint | |
### **๐จ Image Generation** | |
- **Automatic**: High-quality educational images generated automatically | |
- **No Setup Required**: Image generation credentials are pre-configured | |
- **Professional Quality**: Clean, watermark-free educational visuals | |
## ๐ **Why This Demonstrates Real Agent Power** | |
### **๐ Multi-Tool Orchestration** | |
Unlike simple chatbots, our agents actually use real tools: | |
- **Web APIs** for current information | |
- **Image Generation APIs** for visual content | |
- **Multiple LLM providers** for diverse capabilities | |
- **Content Processing Tools** for structured output | |
### **๐ฏ Practical Real-World Application** | |
- **Educators** can generate curriculum quickly | |
- **Students** can create study materials for any topic | |
- **Content Creators** can produce educational resources | |
- **Businesses** can develop training materials | |
### **๐ Scalable Architecture** | |
- **Modular Design** allows easy tool addition | |
- **Provider Agnostic** works with any LLM service | |
- **Error Resilient** handles API failures gracefully | |
- **Performance Optimized** for real-world usage | |
## ๐ **Demo Statistics** | |
- **๐ Courses Generated**: Unlimited topics supported | |
- **๐ Content Volume**: 10,000+ words per complete course | |
- **๐จ Images Created**: Up to 15 educational visuals per course | |
- **๐ง Assessment Items**: 25+ interactive learning elements per course | |
- **โก Generation Time**: 2-5 minutes for complete course | |
- **๐ Language Support**: Multi-language content generation | |
## ๐ก๏ธ **Reliability Features** | |
### **Smart Error Handling** | |
- **Provider Fallback**: Automatically switches if primary LLM fails | |
- **JSON Recovery**: Handles malformed responses gracefully | |
- **Content Validation**: Ensures educational quality standards | |
- **API Resilience**: Continues operation despite individual tool failures | |
### **Quality Assurance** | |
- **Research Verification**: Cross-references multiple sources | |
- **Content Coherence**: Maintains logical lesson progression | |
- **Assessment Alignment**: Quizzes match lesson objectives | |
- **Visual Relevance**: Images enhance specific content points | |
## ๐ฏ **Perfect for Hackathon Judging** | |
### **โ Innovation** | |
- Real multi-tool agent orchestration (not simulated) | |
- Novel educational content generation approach | |
- Creative use of multiple AI services together | |
### **โ Technical Excellence** | |
- Robust error handling and provider fallbacks | |
- Clean, modular architecture | |
- Real-time tool integration | |
- Professional UI/UX design | |
### **โ Practical Impact** | |
- Immediate value for educators and learners | |
- Scalable to any educational domain | |
- Production-ready implementation | |
- Clear business applications | |
### **โ User Experience** | |
- Intuitive interface requiring no technical knowledge | |
- Engaging interactive elements | |
- Professional visual design | |
- Comprehensive feature set | |
## ๐ **Try It Now!** | |
Experience the power of real AI agents working together to create educational content. Generate a course on any topic and watch as our agents research, create, and enhance your learning experience in real-time! | |
**๐ [Launch Course Creator AI](https://huggingface.co/spaces/Agents-MCP-Hackathon/course-creator-ai)** | |
--- | |
## ๐ **Technical Details** | |
- **Framework**: Gradio 4.44.0 | |
- **Language**: Python 3.9+ | |
- **Architecture**: Async multi-agent system | |
- **APIs**: DuckDuckGo, Pollinations, OpenAI, Anthropic, Google | |
- **Deployment**: Hugging Face Spaces | |
- **License**: MIT | |
## ๐ท๏ธ **Tags** | |
`agent-demo-track` `education` `ai-agents` `course-generation` `multi-tool` `web-research` `image-generation` `gradio` `hackathon` | |
--- | |
**Made with โค๏ธ by Pink Pixel** | **Hugging Face Agent Hackathon 2025** | **Track 3: Agentic Demo** |