I swore building it… 
Actually, wasn’t so bad a project, used a Raspberry Pi and a monochrome camera. Described here, under “Power Measurement”:
https://discuss.pixls.us/t/the-quest-for-good-color-4-the-diffraction-grating-shootout/19984
The problem that stopped me was the IR cutoff, eliminated measurements above ~660nm.
Since, I’ve messed around a bit with this:
https://www.sparkfun.com/products/15050
18 channels covering the range 410nm - 940nm. A bit coarse, but you can plot useful power distributions with it. You’d be hard-pressed to read patches with it, however, getting the patch simultaneously to all three sensors. Arduino interface can take advantage of readily available sketches to read it; I did it the hard way on a Raspberry Pi, still haven’t worked out the channel ordering.
This is what I’d like to use:
https://www.digikey.com/en/products/detail/groupgets-llc/BO-HAMA-C12880-V2-SENSOR/14306449
340nm ~ 850nm at 15nm intervals, through a single aperture. Pricey, but well-under the lab-grade devices.
Okay, all that said, probably the easiest-to-use, bang-for-the-buck device is this:
https://calibrite.com/us/product/colorchecker-studio/
The old i1Studio spectrophotometer, originally offered by X-Rite but spun off to a company called Calibrite. In doing so, the price went from ~$470US to ~$569US, serves me right for not getting one earlier. You might find the X-Rite version on Ebay for ~$400US. This thing will read ambient light, patches, 380nm-730nm at 10nm intervals. After this, you’ll be shelling out north of $1000US for a lab-grade device.