IDEA cli

UPDATE Added Ollama Support!

CLI tool to generate short ideas into your Obsidian Vault.
Read the blog post

Config

Add to your .env file or environment vars

VAULT_PATH=/full/path/to/your/vault
OPENAI_TOKEN=openai-token
OPENAI_URL=https://api.openai.com/v1/
OPENAI_MODEL=gpt-3.5-turbo
USE_OLLAMA=YES|NO
OLLAMA_MODEL=llama2:latest
YOUTUBE_TOKEN=youtube-token-not-used-yet

TO DO

• Better error handling.
• Check if there are video's if not, don't generate the content
• Use Official youtube library: google-youtube3
• Integrate in NeoVim (NeoVim plugin!) FEATURE
• Generic config file in ${HOME}/.config/idea-cli/settings.json
• Option for using Ollama instead of OpenAI
• Write tests
• fix the unwrap in the completion
• Strange behaviour in the return of the completion. somehow the string isn't trimmed properly returns a \\u0000-\\u001F
• Optimize code!!!!!
• Make sure not to overwrite a file
• Add tags based upon input
• Write neovim plugin
• Convert returns to Result<T,E>

Build

MacOS

cargo build --target=aarch64-apple-darwin --release

Windows

cargo build --target x86_64-pc-windows-gnu

NeoVim plugin

Check out the Neovim plugin

classDiagram
    class Config {
        +String ollama_model
    }

    class GenerationOptions {
        +GenerationOptions default()
        +GenerationOptions temperature(float)
        +GenerationOptions repeat_penalty(float)
        +GenerationOptions top_k(int)
        +GenerationOptions top_p(float)
    }

    class GenerationRequest {
        +GenerationRequest new(String model, String prompt)
        +GenerationRequest options(GenerationOptions options)
    }

    class Ollama {
        +Ollama default()
        +generate(GenerationRequest request) : Result
    }

    class Result {
        +String response
    }

    class Error {
        +String to_string()
    }

    class Completion {
        +ollama_completion(String prompt, Config cfg) : String
    }

    Config --> Completion : uses
    GenerationOptions --> GenerationRequest : uses
    GenerationRequest --> Ollama : uses
    Ollama --> Result : returns
    Result --> Completion : returns
    Error --> Completion : returns

Loading