Imagine a Personal Second brain living in your computer. You drop your research papers, meeting notes, PDFs, and images into a folder, and it instantly categorizes, reads, and writes an interlinked Personal Wikipedia-style article for you. Best of all? It's 100% private, runs offline, and doesn't send a single byte of your data to the cloud.

(Drop your files in the sidebar and start chatting instantly!)

• 100% Offline & Private: Powered by your local LLM via Ollama. Your personal files stay on your machine.
• Auto-Ingestion Engine: Just drop .pdf, .docx, .txt, or .png (built-in OCR) into the raw/ folder. The Watchdog automatically compiles a comprehensive Wiki page.
• MediaWiki Aesthetics: A beautifully styled web UI that looks and feels like Wikipedia, complete with markdown rendering and bidirectional link styling.
• Chat with your Wiki: Built-in Chatbot that uses local RAG to automatically retrieve pages from your wiki and provide contextual answers.

The project relies on a decoupled two-loop architecture: the Ingestion Engine (Invisible Librarian) and the Web Interface.

graph TD
    User([User]) -->|Uploads PDF, DOCX, TXT| Raw[raw/ folder]
    User -->|Views & Chats| UI[Streamlit App app.py]
    
    subgraph Ingestion Engine ingest.py
    Watchdog[Folder Watchdog] -->|Detects new file| Parser[Document Parser]
    Parser --> |Extracts Text| Compiler[Wiki Compiler]
    Compiler --> |Prompts LLM| Ollama[Local Ollama]
    Ollama --> |Returns Markdown| Compiler
    Compiler --> |Saves .md| Pages[wiki/pages/ folder]
    Compiler --> |Moves original| Archive[archive/ folder]
    end
    
    Raw --> Watchdog
    Pages -.->|Read by| UI
    UI -->|Queries Wiki| Ollama

Before you start, ensure you have the following installed:

1. Python 3.8+
2. Ollama (The local LLM runner)
3. Tesseract OCR (For image support):
   • Mac: brew install tesseract
   • Linux: sudo apt install tesseract-ocr

1. Clone the repository:

git clone https://github.com/aswin-pn/Personal-wiki-using-Gemma-4.git
cd Personal-wiki-using-Gemma-4

2. Set up the Virtual Environment:

python -m venv venv
source venv/bin/activate
pip install -r requirements.txt

3. Pull the Default LLM:

ollama pull gemma4:e4b

4. Configuration: Copy .env.example to .env to configure your preferred LLM.

cp .env.example .env

To run the wiki, you must start both the Librarian and the Web UI in two separate terminals.

Terminal 1: Start the Background Librarian
This script watches for incoming files and converts them into Wiki format automatically.

source venv/bin/activate
python ingest.py

Terminal 2: Start the Web UI
This is your Wikipedia viewer and Chat UI.

source venv/bin/activate
streamlit run app.py

Contributions, issues, and feature requests are welcome! Feel free to check the issues page.

This project is licensed under the MIT License.