Great find, @nickbakeddesign!
Figma’s own HTML → Figma tool can be used in a bookmarklet to grab any page.
Also, bookmarklets are awesome, but fiddly to share. I quietly built a tool last year for sharing and installing bookmarklets with drag and drop:
https://t.co/ZvhduILknV
There's a free bookmarklet that pastes any website into Figma as editable layers. It replaced a paid tool! ($18/m)
it has zero dependency on AI, it runs in your browser and copies any webpage straight to your clipboard as Figma layers using figma mcp.
@nicolaibaaring@benjitaylor Was pretty straightforward!
Had to inject
[data-feedback-toolbar] * { box-sizing: border-box; }
along with it, to fix some overflow issues.
And I set it to click
[class^="styles-module__controlsContent___"]
as soon as it appears in the DOM, so it’s open from the start.
@ninepixelgrid I could never get correlation analysis to work in PostHog, but that seems like something AI could maybe enhance? E.g. separate probable causation from obvious coincidence.
@goo Only thing that comes to mind is prototype sharing in @createwithplay. Which is a clever way to exploit them, but I agree, I can’t say I’ve seen them used as intended, in the wild.
How much nature is lost in Norway – the world’s poster child for sustainable development? More than you think. I got shivers when mapping nature loss for this excellent NRK article. Google-translated article: https://t.co/QnwYmoorXW
Thread about the main findings and methods…
@LeaVerou This, combined with the absence of a required attribute, seems like it might be a sufficient heuristic for a native solution?
No preselected radio button + no required item = the user agent can deselect when clicking the selected item.
@ChetuWilliam I don’t know enough about how the gclid is set to answer that, but there’s nothing about email as a channel that should prevent you from passing the query string parameter itself.