bxss | Inject Blind XSS payloads into custom headers | Security Testing library

I have not, but shall DAST* security test, out of curiosity, an IoT device; Nodemcu esp8266 www server I built. It's showing a HTML page (on a mobile phone for example) that allows to control and interact with a camera module and a A/C relay. With it I can for example show images captured in the camera I even think it has some image recognition built in, and I can switch on and off a relay for electrical current to a light bulb (110/220v A/C power)

Before I start pentest I though I better start thinking of what types of exploits one would be able to find and detect? Which sinister exploits I will be able to find, or rather ought be able to find given a proper pentest exercise? (And if I do not find exploits, my approach to the pentest of the Iot might be wrong)

I ponder it might be a totally pointless exercise since the esp8266 www server (or rather its LUA programming libraries) might not have any security built into it, so basically it is "open doors" and everything with it is unsafe ?

The test report might just conclude what I can foresee be that the the "user input needs to be sanitized"?

Anyone have any idea what such pentest of a generic IoT device generally reports? Maybe it is possible to crash or reset the IoT device? Buffer overruns, XXS, call own code ?

I might use ZAP or Burpsuite or similar DAST security test tool.

• I could of course SAST test it instead, or too, but I think it will be hard to find a static code analyzer for the NodeMCU libraries and NUA scripting language easily ? I found some references here though: https://ieeexplore.ieee.org/abstract/document/8227299 but it seems to be a long read.

So if someone just have a short answer what to expect in a DAST scan/pentest , it would be much appreciated.

Stay safe and secure out there ! Zombieboy