dumbpilot/TODO.md

[x] - in extensions.json add suffix for languages that require it such as css where comments are: /_ stuff _/
[] - test cancel token
[x] - add fill in middle
[x] - add config option to disable the extension
[] - add command to test and query connection to server
[x] - add feedback when waiting response
[] - add a chat window
[] - if the model is an instruct-type add the system prompt to the chat
[x] - add an icon
[] - add a window to quickly modify model configs
[] - sliders for context lengths and generated buttons
[] - option to backup and restore model settings as JSON
[] - decorate ai generated text https://github.com/microsoft/vscode-extension-samples/tree/main/decorator-sample
[] - when trying to use completion when there is an active selection either substitute the selection or use the selection as context instead of the whole file
[] - when in stream mode, print the tokens as they arrive in the status bar as an additional indication that the model is generating and to quickly check if the model has entered a loop
separate llama.cpp server api into a different source file 2023-12-16 16:56:44 +01:00			`[x] - in extensions.json add suffix for languages that require it such as css where comments are: /_ stuff _/`
works tm 2023-11-19 23:37:24 +01:00			`[] - test cancel token`
formatting json files 2023-12-16 19:48:27 +01:00			`[x] - add fill in middle`
some interactivity 2023-11-20 23:17:29 +01:00			`[x] - add config option to disable the extension`
works tm 2023-11-19 23:37:24 +01:00			`[] - add command to test and query connection to server`
more config 2023-11-30 00:00:11 +01:00			`[x] - add feedback when waiting response`
			`[] - add a chat window`
			`[] - if the model is an instruct-type add the system prompt to the chat`
formatting json files 2023-12-16 19:48:27 +01:00			`[x] - add an icon`
more config 2023-11-30 00:00:11 +01:00			`[] - add a window to quickly modify model configs`
added streaming support 2023-12-20 15:42:49 +01:00			`[] - sliders for context lengths and generated buttons`
			`[] - option to backup and restore model settings as JSON`
last chance before changing to OpenAI API 2023-12-14 20:12:43 +01:00			`[] - decorate ai generated text https://github.com/microsoft/vscode-extension-samples/tree/main/decorator-sample`
separate llama.cpp server api into a different source file 2023-12-16 16:56:44 +01:00			`[] - when trying to use completion when there is an active selection either substitute the selection or use the selection as context instead of the whole file`
added streaming support 2023-12-20 15:42:49 +01:00			`[] - when in stream mode, print the tokens as they arrive in the status bar as an additional indication that the model is generating and to quickly check if the model has entered a loop`