ToS read: I've gone through the terms directly, and I don't think this would be prohibited. Remind has a human-in-the-loop setting up these reminders, and uses `claude`'s documented extension points to do its business. I'm just a developer, so take this for what you will.
Notifications: Yeah, I think you could just ask Claude to notify you based on some interest condition. When it writes back into the reminder it's really just for bookkeeping and task continuations.
Recurring: Yes, this works cleanly. Remind marks the Reminder as complete at fire time, letting it roll forward.
CLI: You could have claude just call them, but you're asking if it could fire cli tools directly instead of through `claude`?
Last, I realized I didn't mention how to actually use Remind from `claude` itself: It installs a skill called `/remind` which you can invoke inside the chat.
Based on my research the ivar has been stable since macOS 13. I'll keep things up to date if that ever changes... To your second question, it's mostly for agents to pull icons as they go without interruption or you having to find whatever it chose to replace later. I've found Claude using batch mode after it does a lookup against the list to match a set of icons.
You're right. Just flipped the default to SVG on main. Honest answer: SVG wasn't fully working during early dev, so PDF was the working default and it stuck. Should've revisited.
Thanks! I'm not sure how animations would translate into these exported formats. SF Symbols animations are typically state driven, and that doesn't really translate to a standalone SVG/PNG/PDF. You'd get something that autoplays on loop, or fires once. Let me know if you had something else in mind.
Right, these are licensed by Apple for UI use on Apple platforms only. The README has a "Before you use this" callout up top about this. sfsym is an export tool; what you do with the output is on you.
I’ve had this frustration with a number of subscriptions over the years, but there’s a great workaround—subscribe from the App Store. I get access to the NYT website, and cancellation is one button press away in my subscription settings.
The Skeptics’ Guide To The Universe has been a weekly must-listen for me over the last decade+. They cover “science news, critical thinking, bad science, conspiracies, and controversies”.
I believe his biggest issue with IQ—one that worries me greatly as well—is that it is being used to create a feeling of "otherness" towards populations that can be demarcated out using a single dimensional property: IQ. History shows us this as a strategy to subjugate using other "scientific" methods of classifying people, phrenology being a fairly recent example.
I didn’t have to amend the soil. Thyme actually does well in poor soil conditions! As for planting you’ll have to buy plugs if you want consistency. These varieties if grown from seed won’t necessarily be true to form.
I'm in Albuquerque (zone 7). I'm not sure this is a great kid lawn because while it will hold up to foot traffic, it might get kind of torn up if kids are running around on it a lot.
No, these were all planted as plugs. Each variety of thyme must be propagated via cutting as the traits of that variety are not guaranteed if grown from seed.
Just to shill for bee friendly lawns: I have a creeping thyme lawn that I would recommend entirely. It attracts tons of bees, butterflies, doesn’t need mowing but once a year (to clean up the spent flowers which are lovely in the summer), is drought tolerant (I water once every 2 weeks), and is evergreen to boot!
What would make this ultra killer is to have a VR component. Imagine being able to walk around a design (obviously you'd have to be in a large enough space—basketball court, warehouse, empty parking lot, etc…). You would be able to experience the design at a human scale.
I would imagine that the operators of the current UI were trained on it as well. What the OP posted is certainly an improvement, but could be made bulletproof by showing each choice in context alongside a button that you can't accidentally trigger.
Imagining using this 3-page design makes me feel uncomfortable at step 2 and 3. What if I didn't know what the "test" notification was supposed to look like and wanted to make sure the system took my input correctly? What if it's such a dire emergency that the operator forgets (perhaps from experiencing intense anxiety) which one was selected? I think a stronger approach would be to flatten the design to one page. You could then visually confirm that you've chosen a test/real alert, which alert was chosen, and then see what those options actually produce. In addition, because of the seriousness of sending an alert, I think it would be better to make the "Yes, Send Now" button harder to engage — maybe a slide to send, press and hold to send, or simply give a confirmation dialog after clicking it would help.
You can easily control transformations with it (out of the box), but it's easily extensible to animate/smooth anything that you can control with a number (think sound, brush strokes in a painting app)
Notifications: Yeah, I think you could just ask Claude to notify you based on some interest condition. When it writes back into the reminder it's really just for bookkeeping and task continuations.
Recurring: Yes, this works cleanly. Remind marks the Reminder as complete at fire time, letting it roll forward.
CLI: You could have claude just call them, but you're asking if it could fire cli tools directly instead of through `claude`?
Last, I realized I didn't mention how to actually use Remind from `claude` itself: It installs a skill called `/remind` which you can invoke inside the chat.