nero-dv / generally-pretty-true-assistant Goto Github PK

Describe the solution you'd like
Implement a solution to edit and preview markdown for easier content management. Unsure whether this should be listed as a feature request as there already great third party solutions, and this would increase the complexity for not so much reward.

Is your feature request related to a problem? Please describe.
It is currently not possible to load previous conversations into the chat's context for your next response. This limitation increases the number of tokens someone spends by having to ask the same thing twice if a session has been terminated.

Describe the solution you'd like
A connection to a local (or remote) MongoDB session to store your own personal cache of raw responses. Implementing this will increase complexity, though it is a very useful feature to have. Ideally all connections to the external database would need to be happen securely either through a zero trust solution, SSO, and client/host based TLS handshakes.

Is your feature request related to a problem? Please describe.
The context for the first tab is for chat related requests and is designed to handle a specific type of return from the API.

Describe the solution you'd like
Add additional tab to include intuitive interface for viewing relations between two pieces of text. One idea I've considered is using a spreadsheet type interface along with the same interface that allows Wikipedia editors to view differences between two post revisions.

Is your feature request related to a problem? Please describe.
The only model that can be sent requests is the chatgpt-turbo model

Describe the solution you'd like
Activate the dropdown menu to select additional models. Ideally the model should have the same context of either a chat conversation, or text completion (such as for Codex for code completion)

Is your feature request related to a problem? Please describe.
The context for the first tab is for chat related requests and is designed to handle a specific type of return from the API.

Describe the solution you'd like
Add additional tab to include intuitive interface for uploading audio files to the OpenAI model that handles audio transcription (Whisper)

Is your feature request related to a problem? Please describe.
The context for the first tab is for chat related requests and is designed to handle a specific type of return from the API.

Describe the solution you'd like
Add additional tab to include intuitive interface for uploading images to the OpenAI model that can create (and edit) realistic images and art from a description in natural language.

Describe the bug
The clear history button located in the Edit menu is meant to clear both the conversation and the contents of the history tab.

Expected behavior
Right now, the clear chat option only clears the Chat response window.

Is your feature request related to a problem? Please describe.
At the moment, the application doesn't include the option to set how many responses to include as context for the next response from OpenAI.

Describe the solution you'd like
A selector with the aforementioned action. A stack can be implemented in a toolbar to select the number of chats included for context.

Is your feature request related to a problem? Please describe.
The responses returned by OpenAI are being converted to plain text for parsing. The response contains markers for markdown, which can be displayed if saved to a file ending in MD and opening in a program that can view formatted markdown content.

Describe the solution you'd like
The response window should display things in markdown and include the option to view as plain text. Implementation details are a bit fuzzy due to the way that the response window's child functions work.

Is your feature request related to a problem? Please describe.
At the moment there is no way to select a different API key other than by first having it on hand and entering it through the Set API Key option in the File Menu

Describe the solution you'd like
A manager window to keep track of API keys and their usage

Describe alternatives you've considered
Continue to use the file menu option and use OpenAI's billing dashboard (Why can't I see all my requests on one page instead of by a single date?!...)