I wonder if the type of illumination used for traditional book reading is relevant. I.e. would a bedside lamp with cold light provoke the same effects?
As today, the academic consensus is that Melatonin suppression, on the lower part of the spectrum, is regulated by melanopsin produced by the Intrinsically photosensitive retinal ganglion cells (ipRGCs) -- not blue rods. The photosensitivity of ipRGCs has λ_peek = ~460 (nm).
They've done experiments with rats without cones and rods (but with ipRCS), and they displayed circadian cycles.
Melatonin is also suppressed by lights on higher part of the spectrum, but haven't papers on it.
I've decided in getting this safety glass: https://www.thorlabs.com/thorproduct.cfm?partnumber=LG3A
The dark amber glass from swanwick have similar optical density curve: https://www.swanwicksleep.com/
I choose the safety glass instead of the swanwick glass because I have more confidence that a safety glass in conforming with the provided specification. Since I intend to use it home, I don't care about how it looks.
It you want to customize your own glass, this company sells lens blank of filter that block a somewhat similar wavelength to the other products: https://laserprotection.com/product/laser-pointer-protective...
Zeiss also sells special lens for colorblind that blocks similar spectrum. But it is very hard to find information on how to get them online.
On the other hand, maybe there is some filtering taking place when light is bounced off materials. I.e. blue wavelengths being absorbed by the paper.
A paper[1] funded by NIH published this year showed that ~12 log photons cm⁻² s⁻¹ on λ=460nm is enough to suppress melantonin production.
I've discussed more on sibling thread.
[1] https://journals.sagepub.com/doi/pdf/10.1177/074873041558541...
Almost certainly from my own anecdotal experience. There's a reason why bedroom lights tend to be warm (even though they're often now leds)