It's only been 2 weeks since the last major release, but we're back with another mighty fine update which includes Everywhere mode! Let's dive into all the new goodies.
AI Everywhere
It is now possible to prompt AI from every input field.
To invoke the AI, simply type /ai followed by your query. After pressing
shift, the request will be sent for completion, and the result will be
directly inserted as the value of the input field.
Once activated in the settings, this feature will be available across all inputs fields in the admin area including Bricks, Elementor and Oxygen.
Better data support
Chatbots can now use content from the following page builders in the data section:
- Bricks
- Elementor
- Oxygen
Additionally, Gutenberg blocks are now being converted correctly when using
Post content as a source. Lastly, all content will be stripped of HTML tags,
ensuring a clean final output for the AI to consume.
Meta value modal
When using a source that supports meta keys, it was previously hard to know what
meta values were available under which key. Especially for beginners, dot
notation (post_meta.subtitle) can be confusing. To make this easier, we've
added a modal that lists all available values.
Improved assistants setup
There was a little oversight in the previous release:
Assistants only work when the model is set to a turbo model, so either
gpt-3.5-turbo or gpt-4-turbo. Saving a chatbot with a different model won't
correctly create an assistant. This has been fixed in 1.3. SleekAI will now hide
incompatible models from the setup screen.
Various UX improvements
There are various other miscellaneous improvements and bug fixes in this release:
- OpenAI errors will also be displayed when using streaming
- Messages can be copied to the clipboard
- Only show GPT models in the model dropdown
And that's a wrap! We hope you enjoy these new features and improvements. As always, if you have any questions or feedback, feel free to reach out. Happy chatting! 🚀