ishnaya/langgraph-model

Modular LangGraph Agent (Date, Selenium Scraper, RAG/FAISS, Streamlit UI)

Features

• Date/Time tool (stdlib datetime, timezone-aware)
• Web scraping tool (Selenium, headless)
• RAG tool (FAISS + SentenceTransformers; fallback to scikit-learn TF‑IDF)
• Streamlit chat UI with file upload for RAG (TXT/MD by default; optional PDF/DOCX if libs available)

Quickstart

python -m venv .venv && source .venv/bin/activate  # or .venv\Scripts\activate on Windows
pip install -r requirements.txt
streamlit run chatbot.py

Using File Upload (RAG)

In the Streamlit sidebar, use “Upload documents (RAG)”. Supported out of the box:

• .txt, .md
• If you additionally pip install pypdf python-docx, then .pdf, .docx will also work.

After upload, click “Index uploaded files into RAG”. Then ask questions in the chat – the agent will use the RAG node when your message contains keywords like rag: or when you toggle Force RAG.