Enter your Google API key. Get one here: https://makersuite.google.com/app/apikey
Set default text model to use when not provided in input dictionary.
Possible options: “gemini-pro”
https://platform.openai.com/docs/models/continuous-model-upgrades
Set default text + vision model to use. Possible values: “gemini-pro-vision”
Set model temperature. Value must be between 0 and 2
https://platform.openai.com/docs/api-reference/chat/create#chat-create-temperature
By default, conversations are logged so that they can be accessed using the “View or continue Gemini conversation” companion shortcut. When conversations are logged, Gemini Pro will create a title for the conversation automatically. Should the title be updated with each message to better reflect the conversation content, which slows things down a bit? 0 for no, or 1 to update title on every logged message.
Check for shortcut updates
【Auto-Update Routine】
Checks For Latest Shortcut Version
↳ https://routinehub.co/shortcut/16590/
【Auto-Update Routine Ends Here】
Support → https://www.reddit.com/r/shortcuts/comments/18j4f9p/
【⚙️】
Convert Semantic Number to Decimals
↳ https://routinehub.co/shortcut/16589/
Get date for naming log file
Set default parameter values
Process shortcut input, if provided
Process dictionary input
Empty prompt value, so all user for text/image to use.
Prepare image for sending in API call
Log image
We don’t log the image bytes, so some bookkeeping is done to maintain the messages sent via api versus those saved in logs
Prepare and make API call
Extract returned chatbot response
Update Messages list with chatbot response
Prepare shortcut output
Log conversation
If continuing conversation within shortcut, prepare input for recursive shortcut call