Enterprise Knowledge Search Agent (RAG) 🤖💼

An intelligent, domain-specific semantic search engine and Generative AI agent designed to parse, index, and explain complex enterprise ERP systems terminology, acronyms, and platform modules.

🚀 Features

• Retrieval-Augmented Generation (RAG): Connects a vector database to an LLM to answer complex domain queries dynamically.
• Custom Domain Dictionary: Includes an extensive custom knowledge base (dictionary/) covering Ellucian's product ecosystem:
  • Banner, Colleague, Ethos, Degree Works, ILP, PowerCampus, and more.
• Semantic Acronym Matching: Vectorizes abbreviation definitions and queries using embeddings to find contextually relevant definitions instead of simple keyword matches.
• Flask Web Interface: Provides a conversational UI where users can query terminology and interact with the search agent.

📁 Repository Structure

• agent/
  • vector_store.py — Embeds acronym dictionaries and manages the ChromaDB/SQLite vector database lifecycle.
  • knowledge_agent.py — Orchestrates search queries and formats agent reasoning using OpenAI/LangChain.
  • app.py — Flask server and Web application interface.
  • requirements.txt — Python dependencies.
• dictionary/ — Structured JSON datasets containing domains, acronyms, definitions, and acronym classifications.

🛠️ Tech Stack & Libraries

• Language: Python
• AI/ML: LangChain, OpenAI API, ChromaDB (Vector Store), Sentence-Transformers
• Web Backend: Flask, HTML5, CSS3

📦 Getting Started

1. Installation

Clone the repository and install dependencies:

pip install -r agent/requirements.txt

2. Configure Environment Variables

Set your OpenAI API key in your terminal or a .env file:

export OPENAI_API_KEY="your-api-key"

3. Initialize Vector Database & Start Server

Initialize the knowledge base index and run the Flask application:

python agent/app.py

Open http://127.0.0.1:5000 in your browser to interact with the Knowledge Search Agent.