AI-Powered-Function-Executor

📌 AI-Powered Function Executor

🚀 Overview

AI-Powered Function Executor is a Python-based API service that leverages LLM + RAG (Retrieval-Augmented Generation) to dynamically retrieve and execute automation functions based on user prompts. The system processes user inputs, maps them to predefined functions, and generates executable Python code for invocation.

🎯 Features

✅ Function Registry – Predefined automation functions (e.g., open applications, system monitoring, command execution) ✅ RAG-Based Function Retrieval – Stores function metadata in a vector database and retrieves the best match ✅ Dynamic Code Generation – Generates structured Python scripts for function execution ✅ Context-Aware Execution – Uses session-based memory for better function retrieval ✅ FastAPI-Based API Service – Exposes a REST API for function execution ✅ Extensible & Scalable – Can support user-defined functions in the future

🛠 Tech Stack

• Python (Core language)
• FastAPI (API framework)
• FAISS (Vector database for retrieval)
• Sentence-Transformers (LLM embeddings)
• Uvicorn (ASGI server)
• Psutil (System control & automation)

🔧 Installation & Setup

1️⃣ Clone the Repository

git clone https://github.com/Sim14code/AI-Powered-Function-Executor.git
cd AI-Powered-Function-Executor

2️⃣ Create a Virtual Environment

python -m venv venv
source venv/bin/activate  # For Linux/macOS
venv\Scripts\activate  # For Windows

3️⃣ Install Dependencies

pip install -r requirements.txt

🚀 Running the API

Start the FastAPI server using Uvicorn:

uvicorn api:app --host 0.0.0.0 --port 8000 --reload

🔥 Usage

API Endpoint

POST /execute

Request:

{
  "prompt": "Open calculator"
}

Response:

{
  "function": "open_calculator",
  "code": "<Generated Code Snippet>"
}

📌 Future Enhancements

• ✅ Logging & monitoring for function execution
• ✅ Support for custom user-defined functions
• ✅ Multi-user session handling