DirtyRagger

A working RAG (Retrieval-Augmented Generation) prototype with local document embeddings, vector search, web search, and tool integration.

NOTE: Working on a 2060 RTX with ~6GB of memory!

---

Setup

1. Install dependencies:

pip install -r requirements.txt

2. (Optional) Place your documents in the documents/ folder.

3. (Optional) Build the database:

python build_db.py

This will embed documents into a FAISS vector store.

---

Usage

Run the RAG server

python rag_server.py

Query locally (terminal client)

python rag_client.py

Query via Discord bot

python rag_discord_bot.py

---

Features

• Embeds documents and stores them in a FAISS vector store
• Retrieves relevant knowledge for questions
• Supports web searches for additional context
• Tool integration for advanced RAG workflows

---

Notes

• rag_server.py Runs embedding, reranking, generation, websearch, and tool passes and allows for url queries
• rag_client.py allow for local queries
• rag_discord_bot.py allow for queries and responses through discord