I see the VOC sensor is used to get "CO2 equivalent". I've read that these kinds of sensors are much less accurate for CO2 and not suitable for judging room ventilation:

https://www.environmental-expert.com/articles/co2-vs-voc-sen...

The company does not provide an accuracy level for CO2 readings in the specs.

My experience with particulate sensors has been variable. I have an LED based counter that works quite well at very low particulate levels, and some laser based ones that do not. And another laser sensor (Dylos) that is very accurate at all levels. The accuracy seems to depend a lot on the surrounding electronics and software. While the Canary Clip specs say its particulate sensor is accurate to +/-10ug/m^3 (which at 2.5ug/m^3 per particle means +/- 4 particles in a cubic meter) I wonder what the final _system_ accuracy is.

Has to be led and not laser but still it's a fantastic idea and I wonder how long until air quality sensors start appearing in phones so there is a huge crowdsourced sampling at any event (some high end phones already have temperature and humidity sensors).

They should make it work without a phone though, use the Apple and Garmin APIs to make it work with those watches directly?

I also expect modern devices to have their own downloadable memory at this point without a phone tether needed for sampling (especially at a $100 price point)