
Your Desktop AI Companion
A cross-platform desktop assistant combining a Python backend and a modern web interface for seamless interaction with Google's Gemini model.

Packed with Powerful Features
Direct Gemini Interaction
Chat in real-time with the Gemini 1.5 Flash model for fast and intelligent responses.
Complete Chat History
All conversations are saved locally and are easily accessible right in the sidebar.
Chat Management
Effortlessly create, rename, delete, and clear individual chat sessions to stay organized.
Deep Personalization
Customize themes (Dark, Light, System), API keys, and global system instructions.
Full Markdown Support
AI responses are beautifully formatted with syntax highlighting for code, lists, and more.
Import & Export
Save and load your entire chat history in a portable JSON format for backup or migration.
A Modern Hybrid Architecture
1. Application Launch
The user runs run.py
, which initiates the main application loop in app/main.py
.
2. Dual-Thread Operation
A background thread starts a local Flask server, while the main thread launches a pywebview window.
3. Interface Rendering
The pywebview window displays the user interface by loading the HTML/CSS/JS from the local Flask server.
4. User Interaction
JavaScript sends user messages via fetch
requests to the Flask API endpoint /api/ask_gemini
.
5. AI Communication
Flask processes the request, calls the Gemini API via the official Google library, and awaits a response.
6. Displaying Response
The AI's response is sent back through Flask to the JavaScript frontend, which then renders it in the chat window.
Built With a Robust Tech Stack
Backend
Frontend
GUI & AI
Ready to Dive In?
Get Kiala running on your local machine in just a few steps.
Clone the Repository
Get the source code from GitHub.
git clone https://github.com/Jacqquard/Kiala-v.0.1A.git
Install Dependencies
This will install Flask, pywebview, and other required libraries.
pip install -r requirements.txt
Configure API Key
Create a .env
file in the root and add your key.
GEMINI_API_KEY=YOUR_KEY_HERE
Run the Application
Launch Kiala from your terminal.
python run.py