🏷️ LinkWarden Intelligent Tagger

📌 Overview

🚨 Temporary Solution 🚨

This script was created in response to LinkWarden GitHub Issue #971 as a temporary fix to manually run AI tagging on existing links. It is intended to be a community-driven solution until an official feature is implemented by the LinkWarden team.

Automatically tag your LinkWarden bookmarks using AI-powered tag suggestions! This Python script leverages Ollama's language model to intelligently categorize your saved links based on their content.

⚠️ Important Notes

• This is a community-created temporary solution
• Not an official LinkWarden feature
• May be deprecated once an official solution is implemented
• Use at your own discretion

✨ Features

• 🤖 AI-driven tag suggestions using Ollama
• 📚 Customizable tag list
• 🔍 Smart content analysis
• 🏗️ Flexible configuration options

🛠️ Prerequisites

• Python 3.8+
• LinkWarden instance
• Ollama with Phi3 mini model
• API access to LinkWarden

📦 Installation

1. Clone the repository:

   git clone https://github.com/dalekirkwood/LinkWarden_Manual_AI_Tag
   cd linkwarden-tagger

2. Install dependencies:

   pip install -r requirements.txt

3. Set up your .env file:

   LINKWARDEN_API_KEY=your_api_key_here
   LINKWARDEN_BASE_URL=http://localhost:3002/api/v1  # Optional
   OLLAMA_BASE_URL=http://localhost:11434  # Optional
   SKIP_LINKS_WITH_TAGS=true  # Optional
   

🚀 Usage

1. Customize tags.txt with your preferred tags
2. Run the script:

   python linkwarden_tagger.py

🔧 Configuration

Environment Variables

• LINKWARDEN_API_KEY: Your LinkWarden API key (required)
• LINKWARDEN_BASE_URL: Custom LinkWarden API URL (optional)
• OLLAMA_BASE_URL: Custom Ollama server URL (optional)
• SKIP_LINKS_WITH_TAGS: Skip links that already have tags (optional)

Tag File

Manage your tags in tags.txt. Organize them by categories, one tag per line. Comments start with #.

📋 Requirements

• requests
• python-dotenv
• Ollama with Phi3 mini model installed (see Linkwarden AI Worker Documentation)

📜 License

Distributed under the MIT License. See LICENSE for more information.

🙌 Acknowledgments

• LinkWarden
• Ollama
• Phi3 Model
• Inspiration: LinkWarden Issue #971

---

🌟 Happy Tagging! 🏷️🚀

Disclaimer: This is a community-created tool and is not an official part of the LinkWarden project. Use with caution and always backup your data.