OK, thank you for clearly defining the principle.
I see a number of logical flaws in it. It is an unproven principle that is useless in science. To bring this up in a debate and use this as a basis to say that people who apply scientific reasoning to make a reasonable conclusion to high confidence level are "just claiming" without basis, is inexcusable.
You say:
"Whether a claim is "clearly irrational" or whether it "rises to the level of investigation" is a matter of opinion. The principle is that those who would dismiss it outright are not using a rational litmus test of whether something is worthy of investigation."
According to this there is no rational litmus test whether something is worthy of investigation. It makes an assumption without proof. It states that a person can not use any known facts to make a judgement and labels such a person as irrational.
You say:
"Rationally, there is no harm is allowing even the silliest of experiments to proceed. The silliness of the experiment will manifest as unworthy when, under scrutiny, the experimental rigor falls apart. And frankly, that's is the only rational reason to dismiss a claim/experiment."
This is completely unproven and has evidence against it. There is harm as pointed out by several people (harm to the man in stressing his body, harm to nonscientists who are misled, harm to the pursuit of science because of inefficiency in doing needless experiments, harm to people when taxes and other money is wasted). Also, it assumes that the unworthiness of an experiment is revealed during or after it is done, when the premeditation of the experiment is most crucial. Only a carefully planned and executed experiment can reveal useful data and allowing silly experiments is harmful in itself.
You say:
"So, in principle, unless you actually examine any new claim (i.e. let the experiment proceed), you cannot rationally dismiss it."
You can't dismiss it to 100% certainty, but that is lame argumentation and a useless principle. When a new claim invalidates all other science, you need a good reason to consider the claim. Assuming you find good reasons, you need overwhelming experimental verification to even begin to take it seriously, and silly experiments are not the way to proceed. So, if you want to say the word "dismiss" means 100% dismissal, fine, but it takes a lot of nerve to bring this into a debate in a non-philosophy thread.
In the full light of day, I see this principle for what it is. If no one else does, I'm at a loss for words. If this is the standard to be used at PF, then I must withdraw my call to try to learn some lessons here. There is no point to debate at all.