You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
When I receive a PDF or even an image with appointments, these appointments should be added to the calendar using the assistant #31
If I receive a PDF or even an image with appointments, these appointments should be entered into the calendar using the assistant. I just tested this with ChatGPT and an uploaded image of an A4 sheet with many appointments. In general, it works with ChatGPT.
Describe the solution you'd like
talk or file context with create appointments in calendar
Describe alternatives you've considered
talk or file context with create appointments in calendar
Thanks for this suggestion!
Adding something to a calendar needs to use tools, which is covered by the additional app Context agent. I will move this issue there.
Right now we do not have a provider for image-2-text task types, so it's not possible at the moment to create this, but we will look into it.
Thanks for this suggestion! Adding something to a calendar needs to use tools, which is covered by the additional app Context agent. I will move this issue there.
Right now we do not have a provider for image-2-text task types, so it's not possible at the moment to create this, but we will look into it.
@CoolTuxNet I just checked, both OpenAI and llama take images as input, so adding a task type and provider should be doable. We are happy to receive and help with PRs for this! It would need
Activity
janepie commentedon Jun 2, 2025
Thanks for this suggestion!
Adding something to a calendar needs to use tools, which is covered by the additional app Context agent. I will move this issue there.
Right now we do not have a provider for image-2-text task types, so it's not possible at the moment to create this, but we will look into it.
CoolTuxNet commentedon Jun 2, 2025
That sounds very good. Thank you very much.
janepie commentedon Jun 2, 2025
@CoolTuxNet I just checked, both OpenAI and llama take images as input, so adding a task type and provider should be doable. We are happy to receive and help with PRs for this! It would need
They should be done from top to bottom, but every step would be helpful :)