Python-Code-Assistant

A multi-agent Python application that assists users in programming in Python.
It generates code given a query, validates it, runs it in a sandbox environment, and retrieves relevant documentation to help you understand or expand upon your query.

---

Features

• Plan Generation: Uses the Mistral-Instruct agent to optionally generate a structured JSON plan that is passed to the Coder for better accuracy and context.
• Code Generation: Uses the DeepSeek-Coder agent to generate Python code based on your query.
• Validation: Checks the generated code for errors or inconsistencies.
• Research & Documentation: Fetches relevant documentation or examples to support your query.
• Logging: Tracks all queries and agent interactions for analysis.
• Memory: Will remember the projects and tasks you have been working on (powered by Mem0).

Requirements

• Python 3.11 or higher
• Ollama with deepseek-coder:6.7b, mistral:7b-instruct and for memory purposes phi3:3.8b, nomic-embed-text:latest models
• Additional Python dependencies listed in requirements.txt

⚠️ System requirements may vary depending on the model usage.

---

Installation & Running (Windows)

Option 1: Configure yourself

1. Create a virtual environment

   python -m venv venv

2. Bypass PowerShell execution policy (if needed)

   Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass

3. Activate the virtual environment

   venv\Scripts\Activate

4. Install dependencies

   pip install -r requirements.txt

5. Run the application with Streamlit

   streamlit run app.py

6. Close the application
   Press Ctrl + C in your terminal.

---

Option 2: Faster run option

1. Run the application in your terminal

   python run.py

   This will open a terminal to run the streamlit interface with all packages installed instantly.

2. Close the application
   Press Ctrl + C in your terminal.

Notes

• Make sure your Python interpreter is set to the virtual environment.
• Streamlit must be installed within the virtual environment to run the app successfully.
• Please ensure Ollama is running in the background