An autonomous agent capable of browsing the web (or a simulated database) to collect public information about competitors, synthesize it, and suggest strategic differentiation or repositioning opportunities.
- Complete SWOT Analysis
- PESTEL Analysis
- Porter's Five Forces Analysis
- BCG Matrix
- McKinsey 7S Model
- Competitor Analysis
- Automated LinkedIn Analysis
- Visualization Generation
- Document (TXT for the moment, pdf and other type will be updated soon) Support for RAG Analysis
The project is structured into two main components:
- RESTful API for business analysis
- OpenAI and Tavily integration
- Automated web scraping
- Visualization generation
- RAG support for document analysis
- Modern UI with Tailwind CSS
- Interactive visualizations
- Form management
- Analysis results display
- Docker
Before starting, you need to create a .env file in the project root directory with the following environment variables:
# API Keys
TAVILY_API_KEY=your_tavily_key
OPENAI_API_KEY=your_openai_key
OPENAI_BASE_URL=https://integrate.api.nvidia.com/v1
# LinkedIn Configuration
LINKEDIN_EMAIL=your_linkedin_email
LINKEDIN_PASSWORD=your_linkedin_password- Clone the repository:
git clone https://github.com/kcap02DVT/ghost_analysis.git
cd ghost_analysis
- Create the
.envfile with your API keys:
cp .env.example .env
# Edit the .env file with your API keys- Build and start the containers:
# Build the images
docker-compose build
# Start the services
docker-compose up -d- Verify that containers are running:
docker-compose ps- Access the application:
- Frontend: http://localhost:8080
- Backend API: http://localhost:8000/analyze
Use the confidential.txt file as input for the analysis. This file contains sample data that you can use to test the various analysis features.
- FastAPI
- Uvicorn
- LangChain
- OpenAI API
- Tavily Search API
- Selenium
- BeautifulSoup
- Matplotlib
- React
- TypeScript
- Vite
- Tailwind CSS
- Headless UI
- Lucide Icons
For any questions or suggestions, please open an issue on GitHub.