AISE AI Agent

Dieses Projekt ist Teil der AISE-Aufgabe.

Setup

Virtual Environment erstellen

# Windows
python -m venv venv
.\venv\Scripts\activate

# macOS/Linux
python3 -m venv venv
source venv/bin/activate

Dependencies installieren

pip install -r requirements.txt

Projekt ausführen

# Streamlit App starten
streamlit run frontend/app.py

Tests ausführen

pytest tests/

Projektstruktur

AISE_AIAgent/
├── frontend/                 # Streamlit UI
│   ├── app.py               # Main Streamlit application
│   ├── files_nav.py         # Files Navigation component
│   ├── code_editor.py       # Code Editor component
│   └── chat_ui.py           # Chat UI component
│
├── backend/                  # Backend Logic
│   ├── managers/            # Business logic managers
│   │   ├── file_manager.py      # File operations
│   │   ├── chat_manager.py      # Chat handling
│   │   ├── system_prompter.py   # System prompts
│   │   ├── search_manager.py    # Search functionality
│   │   ├── execution_engine.py  # Code execution
│   │   └── debug_logger.py      # Logging & debugging
│   └── utils/               # Utility functions
│
├── tests/                    # Unit tests
├── data/                     # Data storage
├── docs/                     # Documentation
├── config/                   # Configuration files
│
├── requirements.txt          # Python dependencies
├── README.md                 # This file
└── .gitignore               # Git exclusions

Architecture

Das Projekt folgt einer Frontend-Backend Split Architektur:

• Frontend (Streamlit): User Interface mit Files, Code Editor und Chat
• Backend Modules: Spezialisierte Manager-Klassen für einzelne Aufgaben
  • FileManager: Datei-Operationen
  • ChatManager: Chat-Logik
  • SystemPrompter: Systemanweisungen
  • SearchManager: Suche
  • ExecutionEngine: Code-Ausführung
  • DebugLogger: Logging & Debugging