A natural language processing system that uses LangGraph for intent-based routing to handle various types of requests.
- Question Answering: Get responses to your inquiries
- Code Operations: Generate and edit code across different programming languages
- Text Operations: Create and modify text content including creative writing
├── Dockerfile
├── LICENSE
├── README.md
├── agent
│ ├── __init__.py
│ ├── config.py
│ ├── helper.py
│ ├── main.py
│ ├── nodes
│ │ ├── __init__.py
│ │ ├── analyzer.py
│ │ ├── code_processor.py
│ │ ├── intent_classifier.py
│ │ ├── question_handler.py
│ │ ├── response_generator.py
│ │ └── text_processor.py
│ └── utils
│ ├── __init__.py
│ ├── file_utils.py
│ ├── llm.py
│ ├── output_manager.py
│ ├── router.py
│ └── state.py
├── configs.yaml
├── docker-compose.yml
├── images
│ └── graph.png
├── requirements.txt
└── run.sh # runner
- Python 3.10 or higher
- pip package manager
- Clone the repository
git clone [email protected]:GoshKolotyan/Agentic-System.git
cd Agentic-System- Create a virtual environment
python -m venv agent-venv- Activate the virtual environment
- On Windows:
agent-venv\Scripts\activate- On macOS/Linux:
source agent-venv/bin/activate- Install dependencies
pip install -r requirements.txt- Build and run using Docker Compose
docker-compose up- For detached mode and interactive access
docker-compose up -d
docker attach synopsis-langgraph-processor-1Note: Add your OpenAI API key as an environment variable
export OPENAI_API_KEY='your-api-key'chmod +x run.sh
./run.sh- Question Answering
What is DNA?
- Code Generation
create a C loop that counts from 1 to 10 and save in loop.c
-
Code Editing
# Fix the bugs in this Python script: dec greet(name): if name == "": return "Hello, stranger!" elie: return "Hello, " + name + "!" # Test the function print(greet("Bob")) print(greet(""))
-
Text Generation
Generate a short novel about Bernhard Riemann
- Text Editing
Edit this text for me: The quik brown foxes jumps over the lasy dog and the moon is made of green cheese.
