CrunchyFritos I've run into a few examples of this where the Research Assistant in Expert mode is unable to extract the names of text sections that are obvious in both the text and the html. Even when I specify to list all headings tagged with <h2> for example, the Assistant seems unable to do so. Site: https://daringfireball.net/2024/01/the_vision_pro Thread: https://kagi.com/assistant?mode=7&sub_mode=2&q=https%3A%2F%2Fdaringfireball.net%2F2024%2F01%2Fthe_vision_pro&thread=YKCwlwpvVNn6IhywAeiGlRn6udV4ClDr Same thing here: https://stratechery.com/2024/the-apple-vision-pro/ I expect a list of all the section headings as indicated in the text.
CrunchyFritos FWIW the output I was trying to achieve was to just create a "table of contents" that extracts all the headings and sub-headings from a given long-form piece.
