Agent_project

StoryTeller is a high-performance AI Agentic system built using LangGraph to orchestrate creative workflows. The system autonomously manages a multi-node pipeline: it analyzes a user's initial idea, generates a comprehensive creative narrative, and produces a corresponding high-fidelity visual asset.

Project Structure

The project follows a modular design for scalability:

• Streamlit_app.py: The main interface for user interaction and workflow visualization.
• graph_builder.py: Defines the LangGraph architecture and node connections.
• orchestrator.py: The "brain" of the agent that analyzes requirements and prepares prompts.
• story_writer.py: Specialized node focused on narrative drafting and creative writing.
• tools.py: Contains the generate_image tool for external API integrations.
• state.py: Defines the StoryState schema used to pass data between nodes.

---

Installation & Setup

1. Clone the Repository

git clone https://github.com/SaraMohamed-121/Agent_project.git
cd Agent_project

2. Set Up Virtual Environment

It is recommended to use a virtual environment to manage dependencies:

python -m venv agent_env
agent_env\Scripts\activate

3. Install Dependencies

pip install -r requirements.txt

4. Configuration

Create a .env file in the root directory and add your API credentials:

GOOGLE_API_KEY=your_api_key_here

5. Run the Application

streamlit run Streamlit_app.py

---