ReviewAI is an AI-powered tool that searches, crawls, and analyzes online reviews of movies, video games, and books.
It aggregates critic opinions, summarizes pros and cons, and provides nuanced recommendations tailored to what the user may be looking for — without giving a simple "yes" or "no" answer.
- 🔍 Web Search & Crawling – Automatically finds relevant review and summary URLs using APIs like Serper.dev.
- 📄 Content Scraping – Extracts review text from trusted domains (e.g., IGN, Rotten Tomatoes, Metacritic, Gamespot).
- 🧠 AI Analysis – Uses LLMs to summarize consensus, highlight strengths/weaknesses, and make recommendations.
- 💾 Caching System – Implements persistent semantic caching with ChromaDB to avoid redundant requests and speed up results.
- ⚡ Extensible Graph Workflow – Built with LangGraph for modular processing steps.
- Python 3.12+
- LangChain & LangGraph
- ChromaDB – persistent vector storage for semantic cache
- Serper.dev – Google Search API
- Crawl4Ai – web crawling & scraping
- OpenAI GPT Models for LLM-powered analysis
- Clone the repo:
git clone https://github.com/yourusername/media-review-assistant.git cd media-review-assistant - Install dependencies
pip install -r requirements.txt
- Setup environment varibales in a .env file
OPENAI_API_KEY=your_openai_api_key SERPER_API_KEY=your_serper_api_key
- Run the application
streamlit run app.py