RAGnificent 1.0

A simple RAG chatbot supporting multiple LLM providers

RAGnificent - Your Magnificent RAG-Powered Chatbot Toolkit

RAGnificent is a Python package that enables developers to quickly build powerful chatbots with seamless tool integration and Retrieval-Augmented Generation (RAG) capabilities, supporting any OpenAI-compatible LLM.

Why RAGnificent?

• LLM Agnostic - Works with Groq, OpenAI, Gemini, and any OpenAI-compatible API
• Easy Tool Integration - Add custom functions as tools with minimal code
• Flexible Configuration - Support for both cloud and self-hosted LLMs
• Conversation Management - Efficient short-term memory management with summarization technique
• Prompt Customization - Flexible system and summary prompts
• Lightweight - Minimal dependencies, maximum functionality

Installation

1. Download the package from GitHub
2. Install using pip:

pip install path/to/RAGnificent-<version>-py3-none-any.whl

Quick Start

from RAGnificent import ChatAI, AgentParams
import os


def add(x: int, y: int) -> int:
    """Add two numbers together."""
    return x + y


tools = [add]

# For OpenAI-compatible endpoints
rag = ChatAI()
chatbot = rag.initiate_chatbot(
    params=AgentParams(
        model="gpt-3.5-turbo",  # Or any other model
        api_key="your_api_key",
        base_url="https://api.openai.com/v1",  # Or your custom endpoint
        system_prompt="You are a helpful AI assistant.",
        summary_prompt="Summarize the conversation concisely.",
        thread_id='1',
        tools=tools,  # Optional
        temperature=0.7  # Optional
    )
)

while True:
    user_input = input("You (q to quit): ")
    if user_input.lower() == 'q':
        break
    response = chatbot.run(messages=user_input)
    print("AI:", response)

Configuration Options

AgentParams

Parameter	Type	Description	Required
model	str	Model name (e.g. "gpt-3.5-turbo")	Yes
api_key	str	Your API key	Yes
base_url	str	API base URL (default: OpenAI)	No
system_prompt	str	Initial system prompt	Yes
summary_prompt	str	Prompt for conversation summaries	Yes
thread_id	str	Conversation thread identifier	Yes
user_information	dict	User metadata for personalization	No
tools	list[callable]	Custom tools/functions to integrate	No

Supported LLM Providers

• OpenAI (including Azure OpenAI)
• Groq
• Gemini
• Any OpenAI-compatible API (LocalAI, vLLM, etc.)
• Anthropic Claude (via OpenAI compatibility layer)

Adding Custom Tools

def multiply(a: int, b: int) -> int:
    """Multiply two numbers together."""
    return a * b

def get_weather(city: str) -> str:
    """Get current weather for a given city."""
    return f"Weather in {city}: Sunny"

tools = [multiply, get_weather]

Best Practices

1. Use environment variables for API keys
2. Include clear docstrings for your tools
3. Use type hints for better tool understanding
4. Keep system prompts concise but descriptive
5. Handle sensitive user information appropriately

License

RAGnificent is licensed under the RAGnificent-License:

Copyright (c) 2025 [K. M. Abul Farhad-Ibn-Alam]

Permission is hereby granted to any person obtaining a copy of this software
and associated documentation files (the "Software") to use, modify, and distribute
the Software for any purpose, subject to the following conditions:

1. Redistributions must retain this copyright notice.
2. Commercial use requires written permission from the author.
3. The author is not liable for any damages arising from Software use.

All rights not expressly granted are reserved by the author.