LangGraph + Streamlit Dev Agent

This project is a multi-agent developer assistant built with LangGraph and exposed through a Streamlit web UI.

You describe the app you want in natural language, and the agents will:

1. Plan the application (features, tech stack, file structure)
2. Architect the implementation as explicit tasks
3. Code the actual project files inside the generated_project/ folder

The Streamlit interface lets you:

• Enter your prompt
• Inspect the generated plan and task plan
• Browse and view the generated files directly in the browser
• Download the complete generated project as a ZIP file

---

Project Structure

.
├── app.py           # Streamlit web UI
├── graph.py         # LangGraph setup and compiled agent
├── prompts.py       # Planner, Architect, and Coder prompts
├── states.py        # Pydantic models for Plan, TaskPlan, CoderState
├── tools.py         # Tools for reading/writing files and running commands
├── generated_project/  # Auto-generated project files (created at runtime)
└── .env             # Environment variables (GROQ API key)

Requirements

• Python: 3.10 or higher is recommended
• A GROQ Cloud API key

Installation

1. Clone the repository:

   git clone https://github.com/DasunThathsara/multi-agent-application-builder.git
   cd multi-agent-application-builder

2. Create and activate a virtual environment:

   python -m venv venv
   venv\Scripts\activate

3. Install the required packages:

   pip install -U pip
   
   pip install \
   python-dotenv \
   streamlit \
   langgraph \
   langchain-core \
   langchain-groq \
   pydantic

Environment Setup

1. Create a .env file in the project root with your GROQ API key:

   GROQ_API_KEY="your_groq_cloud_api_key_here"

Running the Application

1. Start the Streamlit app:

   streamlit run app.py

2. Open your web browser and navigate to http://localhost:8501 to access