I don't know how they do it these days, but I think it's probably about the same. This is in the context of military manufacturing in the US in the 1980's when I was involved in areas such as military communications equipment and other military electronics.
In the "old" days a device or product had to have an "Environmental Design Criteria Test Plan" written specifically for it. These were typically a sub-part of a larger document called a "Quality Assurance Project Plan" (QAPP). I used to write them. An EDCTP contained all the tests the "product" had to go through. What tests and things like vibration profiles, temperature ramp rates, duration of each test, and all the other tests and applicable criteria were all in the EDCTP. The tests themselves were usually from "Mil Specs" such as Mil-Std-810 which contained a number of specific types of tests, or a derivation of a standard Mil Spec test, but other specific tests were often defined within the specification for the DUT. Note that the DUT could be a component of a finished product or the finished product its self. I really can't remember any of the specs off hand. I worked with a lot of them. Many of them are no longer in existence.
The EDCTP was what was followed along with referenced specifications & tests, as well as anything applicable in the QAPP. There were no "check lists of any kind other than some general ones used for "standard" test reports for some tests. Many times test reports were specific to the DUT, though.
So in short, the EDCTP was the "Bible" along with referenced tests.
In reviewing a test plan the only thing I did was cross reference the QAPP, the EDCTP and any other referenced standards and specifications to ensure the test plan was in conformance with all the requirements. I can't really tell you anything specific with respect to what to look for other than that.
As I mentioned, part of the process was having experience/training to "know" what to look for. Sometimes part of that was to understand what the purpose of the test was to be able to assess whether the test parameters "make sense". The purpose of the test was usually defined to some degree in the EDCTP.
The biggest problem for me back then was having qualified technicians performing the tests as written. I remember one test technician who screwed up and programmed the controller in a temperature chamber in degrees C when the test called out a temperature in degrees F. Needless to say when the DUT started smoking we knew something was very, very wrong.
I have attached an old "demo" (unfinished/in-process) EDCTP I wrote years ago. It may help you in thinking about this type of environmental testing.
.