Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Feature requests: exporting command, auto-minimize, custom command in Prompt Anywhere and Chat, and plugin documentation #4

Closed
irosyadi opened this issue Aug 1, 2024 · 14 comments
Labels
documentation Improvements or additions to documentation enhancement New feature or request

Comments

@irosyadi
Copy link

irosyadi commented Aug 1, 2024

Hi, I am a big fan of Witsy, and it has become my daily tool of choice. Previously, I used a custom Autohotkey script for my tasks, but Witsy replaced it due to its numerous features. The "AI Command" feature is the one I use the most. However, I have some feature requests:

  1. An option to export/import "AI Command". I have several custom "AI Command" settings that I want to back up and export to another computer without rewriting them.
  2. An option to automatically minimize Witsy when it starts with Windows login.
  3. The ability to use Custom "AI Command" with the "Prompt Anywhere" command. This would enable faster execution of frequently used prompts. For example, I often create dummy JSON files. I could select the custom command for JSON creation and then just prompt the content in "Prompt Anywhere".
  4. Also, the ability to customize prompts (from https://github.com/f/awesome-chatgpt-prompts) in the main chat window since I have some custom prompts that better meet my requirements than the provided ones.
  5. I am not sure how to get the plugins to work. I tried all of them with no success. I searched for documentation but it seems to be unavailable.
  6. Witsy version and update notification. I don't know the Witsy version I am using or whether it needs to be updated to a new version.

Witsy has been a fantastic tool for me and I appreciate you making it so great.
Thank you very much.

@nbonamy
Copy link
Owner

nbonamy commented Aug 1, 2024

Hi @irosyadi and thanks for the positive feedback!

  1. That is easy. Meanwhile you can look in the application configuration folder (https://www.electronjs.org/docs/latest/api/app#appgetpathname) and copy the commands.json file

  2. Tried to do that with Electron but it seems MacOS is preventing to do that in latest versions. Will take another look.

  3. OK. Will think about that.

  4. Same

  5. Plugins should work automatically as long as you have configured them. Ask GPT to summarize a URL and it should fetch the content automatically for instance

  6. Is obvious on the Mac version (About and Check for Update are in the main menu). Will run Witsy on Windows and check!

@nbonamy nbonamy added documentation Improvements or additions to documentation enhancement New feature or request labels Aug 1, 2024
@irosyadi
Copy link
Author

irosyadi commented Aug 2, 2024

Thanks!

  1. I found the command.json file in %APPDATA%\Roaming\Witsy.
  2. Oh okay, I understand how the plugins work by adjusting the prompt.
  3. Yes, I use Witsy on Windows and couldn't find any About or Update Check menus.

@nbonamy
Copy link
Owner

nbonamy commented Aug 19, 2024

Just released 1.8.2:

  • About Witsy in Windows tray icon menu. Check for updates need more work
  • Import/Export commands

@nbonamy
Copy link
Owner

nbonamy commented Sep 2, 2024

Just released 1.10.0:

  • Customized Prompts renamed to Experts
  • Experts are now editable / exportable / importable
  • Experts are now available in Prompt Anywhere

@irosyadi
Copy link
Author

irosyadi commented Sep 3, 2024

Wow, thank you very much. They pretty much meet my workflow when using Witsy.

@nbonamy
Copy link
Owner

nbonamy commented Sep 10, 2024

Closing this issue and opening two of the remaining things:

  • Automatically minimize on startup
  • Auto-update for Windows

@nbonamy nbonamy closed this as completed Sep 10, 2024
@senovr
Copy link

senovr commented Nov 19, 2024

@nbonamy , follow-up question on this already-closed issue.

Can you please confirm, that when we are speaking about search with tavily api key, you have to provide url where to search?

So, it is not the same experience like in perplexity - where you just type a question and it search first, and summarizes it with LLM?

@nbonamy
Copy link
Owner

nbonamy commented Nov 19, 2024

Hi. Not you don't. Most models will decide "by themselves" if an Internet search is required to answer your question or not. If they "decide" not to, you can explicitly ask them in the prompt

Untitled.mov

@senovr
Copy link

senovr commented Nov 19, 2024

Hmm, just tried it with ollama-hosted mistral-nemo, and can not make it to use internet search... my test case was simple - who is president-elect on 2024 elections in US, and mistral keep answering (and hallucinating from time to time) i dont know, i can not look into future. This happens even when I mention - use internet search for answering this question.

@nbonamy
Copy link
Owner

nbonamy commented Nov 19, 2024

Ollama is not supported because of ollama/ollama-js#123.

@senovr
Copy link

senovr commented Nov 19, 2024

ok, thanks for clarification. Will follow this issue to track progress.

@nbonamy
Copy link
Owner

nbonamy commented Nov 28, 2024

Available in 1.26. Probably some tweaking...

@senovr
Copy link

senovr commented Nov 29, 2024 via email

@nbonamy
Copy link
Owner

nbonamy commented Nov 29, 2024

Created #20 to discuss this further

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
documentation Improvements or additions to documentation enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

3 participants