300F is pretty hot for flue gas, so I'm going to assume you're going worst case on the high side.
There are two options that come to mind right off hand, depending on your skill set and which route you want to to go.
There's the K-type thermocouple sensor, which is along the same lines as the Rubicon's one that you referred to. They generally go WAY up in temp, 750c being their max (about 900f). If you wanted to go that route, I'd suggest Adafruits K-Type breakout board ....
https://www.adafruit.com/product/1778 and of course you'd need the sensor itself ...
https://www.adafruit.com/product/270 - total will run you about $30 - However, that will only give you a ANALOG DC Voltage preportional to the temperature (formula in the product listing) - Then you'll need to get that analog data ( like 1.73v ) into domoticz. Easily done, depending on personal skill set. I'd use a arduino to put the data out to Domoticz, but that's just because that's what I have on hand right here, there are many different ways of doing it, and prices vary, and a arduino is about ~$30, total of about $60. There will need to be some programming and behind the scenes 'magic' coding to get everything talking to each other, but nothing too hard or complicated.
The second option that comes to mind right away is a high temp Dallas DS18B20 -
https://www.adafruit.com/product/3846 This is rated up to 257f, but if care is taken, it could 'proportionally' measuse temperatures that are higher, (once you figure out the offset). This is by far the cheapest route, with the sensor running $21, and if you're running Domoticz on a raspberry pi, that will cut your overhead cost WAY down, otherwise, I'd suggest getting a raspberry pi that can directly interface this sensor (along with a 4.7k resistor) and let the pi directly feed Domoticz the data for you. Cut's out the middle man of the K-Type amplifier, and the arduino to convert the analog data.
In regards to the second option, when I mentioned the 'offset' - what I mean is that I personally would maybe add a thermal barrier between the flue pipe and the sensor, just as a 'buffer'. It could be something as trivial as a metal cloths hanger holding the sensor 1/2 inch away from the pipe. Then, to figure out what the offset would be, take the reading of the flue pipe itself by another means, such as a laser surface meter (temperature gun) and compare that to the reading from sensor. Let say the surface temp of the pipe checked with the gun is 150f and the DS18B20 said it was picking up the radiant heat at 100f, so -50f offset. Now, in a prefect world, since the thermal transfer properties of the pipe didnt change, and the distance didnt change, then when the surface temp was 200f then the sensor SHOULD read about 150f, and 300f should be in ballpark of 250f. Then, in domoticz, you'd just throw in a -50 offset, so when domoticz saw 150f, it would actually show 200f. Your mileage and results may vary. Nobody correct me on humidity or air density vs radiant heat!
IMHO, this would be the easiest, safest and most cost efficient way of doing it. Unless you need a contact to turn on at EXACTLY 183 degrees, or start a blower fan at EXACTLY 213 degrees, then I personally would go with option 2.
I currently run the cheaper (non-HIGH TEMP) sensors literally IN my plenum about 2 feet directly after the heat exchanger right before the takeoff ductwork, and I've never seen temperatures above 180 or 190. Although this is different than measuring the flue pipe, i'm just saying with same pre-planning and adapting the technology to work with your usage requirements, it IS doable without any problem.
Hope this helps, or at least points you in the right direction of getting started.
Richard
