memgraph / odin

MIT License
580 stars 14 forks source link

react typescript styledcomponents

obsidian docker

ODIN - Obsidian Driven Information Network

Table of contents

  1. Features
  2. Disclaimer
  3. Prerequisites
  4. Installation
    1. Docker
    2. Manual

Features

odin

Most features are accessible through the Graph Prompt view button in the menu opened by clicking the Expand button in the right upper corner of Obsidian.

  1. Prompt Bar for LLM Queries

odin prompt bar

  1. Graph Visualization
  1. Dropdown Menu Functions

odin dropdown menu

Right click on the highlighted text in the editor to access the following features:

Disclaimer

Warning It is recommended that you have access to GPT-4 via the OpenAI API. GPT-3.5 will probably fail to make correct knowledge graphs from your data. Since we still don't have access to GPT-4 OpenAI API, although we made our account a month ago and generated >1$ in billing a week ago, the init_repo, update_file and add_file endpoints are still untested. We initialized knowledge graphs manually, through ChatGPT. Here be dragons.

Prerequisites

Before you begin, make sure you have the following:

Installation

  1. Download the Plugin:

    • Clone the repository inside the plugins folder (your_vault/.obsidian/plugins) using Git:
      git clone https://github.com/memgraph/odin.git
  2. Install Dependencies and Start the Plugin:

    • Open your terminal or command prompt.
    • Navigate to the plugin's root directory:
      cd odin
    • You have the option to install ODIN using Docker, which will automatically install, set up and run the Memgraph database, the backend, and frontend components, or you can manually run the project locally for a more customized setup or if you already have Memgraph up and running.

      Docker installation

      Before you start, make sure you have a running Docker instance and Docker compose installed.

      1. You will need to create a .env file inside the ODIN folder with your OpenAI API key to access the app features. It should look like this:

        OPENAI_API_KEY=YOUR_API_KEY
        LLM_MODEL_NAME=gpt-4 # try with other models at your own risk 

        Where YOUR_API_KEY is a key you can get here.

      2. Run:

        docker compose up

      It will take up to ten minutes to download and run all dependencies. Now, that you have ODIN successfully installed, you can go to the next step.

      Manual installation

      Make sure you have Node.js version 14 or above and npm installed.

      1. Install the required Node.js dependencies:

        npm install
      2. Start the development build:

        npm run dev

      You now have the app frontend up and running.

      1. You will also need to run the Memgraph database and the application backend by following the installation steps for BOR - backend for Obsidian and Rune.
  3. Disable Restricted Mode:

    • In the Obsidian settings, navigate to "Options."
    • Click on the "Community plugins" tab.
    • Click the "Turn on community plugins" button.
  4. Enable the Plugin:

    • If you don't see ODIN in the list, try reloading Obsidian.
    • Navigate back to the "Community plugins" section in the Obsidian settings.
    • Find "ODIN" in the list of plugins.
    • Toggle the switch next to the plugin name to enable it.
  5. Plugin Usage:

    • The ODIN plugin is now installed and active. You can access its features through the Obsidian interface.