delight2004/runyankole_mcp_server

# 🗣️ Runyakole/Rukiga Language Tutor Server ## 🚀 Overview This project is a **Model Context Protocol (MCP)** server designed to be a personalized AI tutor for learning **Runyakole/Rukiga**. It uses your custom language data to provide highly accurate and contextualized explanations via **Google Gemini**, tackling the scarcity of digital resources for less commonly spoken languages. --- ## ✨ Features - 🔤 **Customizable Knowledge Base**: Load your own Runyakole/Rukiga dictionary, grammar, and phrases from `runyakole_rukiga_data.txt`. - 🧠 **Intelligent Context Management**: Combines your language data with user queries for targeted LLM prompts. - ⚙️ **Google Gemini Integration**: Leverages Gemini for AI-powered explanations and learning. - 🔌 **RESTful API**: Clear endpoints for easy integration with front-end apps or mobile clients. - 🔐 **Secure API Key Handling**: Uses `.env` for storing your Gemini API key safely. - ♻️ **Reloadable Data**: Update your language data without restarting the server. --- ## 🛠️ Technologies - Python 3.x - Flask (Web Framework) - Flask-CORS (CORS handling) - Requests (HTTP client) - python-dotenv (Environment variables) - Google Gemini API (LLM) --- ## ⚙️ Setup & Installation ### 1. Clone the Repository ```bash git clone github.com cd runyankole_mcp_server 2. Install Dependencies pip install Flask Flask-CORS requests python-dotenv # OR pip install -r requirements.txt 3. Get Gemini API Key Sign in to Google AI Studio Generate and copy your Gemini API key. 4. Configure Environment Variable Create a .env file in the project root: GEMINI_API_KEY="YOUR_ACTUAL_GEMINI_API_KEY_HERE" > ✅ Note: .env is already in .gitignore and will not be pushed to GitHub. 5. Create Language Knowledge Base Create runyakole_rukiga_data.txt in the root directory and add your custom language knowledge. Example: --- Runyakole/Rukiga Language Knowledge Base --- Greetings: - Hello (singular): Orair …