https://github.com/emredeveloper/kurum-asistan-chatbot
A modern and functional chat assistant based on Local LLM that streamlines internal processes. Weather, internal information, support requests, document uploads, and more, all on one screen!
https://github.com/emredeveloper/kurum-asistan-chatbot
ai chatbot llm python
Last synced: 6 months ago
JSON representation
A modern and functional chat assistant based on Local LLM that streamlines internal processes. Weather, internal information, support requests, document uploads, and more, all on one screen!
- Host: GitHub
- URL: https://github.com/emredeveloper/kurum-asistan-chatbot
- Owner: emredeveloper
- Created: 2025-06-13T21:53:10.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-10-19T10:48:54.000Z (9 months ago)
- Last Synced: 2025-10-19T20:58:03.048Z (9 months ago)
- Topics: ai, chatbot, llm, python
- Language: Python
- Homepage:
- Size: 151 KB
- Stars: 12
- Watchers: 0
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Internal Smart Assistant Chatbot
A modern and functional LLM-based (local LM Studio/OpenAI compatible API) chatbot that simplifies internal processes. Weather updates, internal knowledge, support tickets, document uploads, and more — all in one screen!
## Features
* 🤖 **LLM-powered natural language chat** (local model via LM Studio/OpenAI compatible API)
* 🔗 **Multi-tool/function chaining**: weather, internal knowledge, support tickets, document upload
* 🏢 **Internal knowledge base**: FAQs, procedures, policies
* 🌤️ **Weather query** (OpenWeatherMap API)
* 💼 **Support ticket creation** (department, description, urgency, category)
* 🗂️ **Dashboard**: query history, weather history, support tickets, uploaded reports
* 📄 **Word/PDF report upload**: attach and manage files from both the chat screen and dashboard
* 🌗 **Dark/Light theme** (persistent with localStorage)
* 📱 **Modern, responsive, and mobile-friendly UI**
* 🛡️ **Secure API key management with .env**
## Installation
1. Clone the repository and navigate into the directory.
2. Create and activate a virtual environment:
```bash
python -m venv .venv
# Windows: .venv\Scripts\activate
# Linux/Mac: source .venv/bin/activate
```
3. Install requirements:
```bash
pip install -r requirements.txt
```
4. Add your OpenWeatherMap API key to the `.env` file:
```
OPENWEATHER_API_KEY=YOUR_API_KEY
```
5. Use LM Studio or Ollama (both can serve as an OpenAI-compatible server):
* **LM Studio**:
* Start a model and enable the OpenAI Compatible Server (e.g. `http://localhost:1234/v1`).
* **Ollama (optional)**:
* Run a model such as `ollama run qwen3:8b` and use an OpenAI-compatible proxy.
* Example `.env`:
```
LM_STUDIO_BASE_URL=http://localhost:1234/v1
LM_STUDIO_MODEL=openai/gpt-oss-20b
# LM_STUDIO_API_KEY=optional
OPENWEATHER_API_KEY=YOUR_API_KEY
```
6. Start the application:
```bash
python app.py
```
7. Open `http://localhost:5000` in your browser.
## Usage
* **Chat screen:** Type and send your message — ask about weather, internal knowledge, support, or document uploads naturally.
* **Report upload:** Add Word/PDF files from the chat screen or dashboard, manage and download them on the dashboard.
* **Dashboard:** View query history, weather history, support tickets, and uploaded reports.
* **Theme:** Switch between dark and light mode with the button in the top-right corner.
## Notes
* All data is temporarily stored in memory; user identity support can be added.
* Departments, knowledge base, and tool chains can be easily customized.
* Supports advanced multi-tool/function chaining with LLM integration.
Developer: emredeveloper