On topic or not
The vast majority, if not all questions here, involves testing ideas against rules. Going through the ones currently at the top of the feed;
Idea of adrenaline-based combustion; tested against biology and chemistry
Idea of a black hole moon detectable via gravitational lensing; tested against physics
Idea of economically independent space stations; tested against economics and some sociology
So sets of rules are often branches of science, which are essentially categories of rules and models following our understanding of reality. Then the internal-consistency questions have the addition of testing the idea against itself, because the idea can be used to make up rules.
In some of these questions the idea is fully thought out and the question is for validation/consistency, in others it is an idea with a specific gap (the problem to solve), and in still others it's a mere concept that needs narrowing down in order to create the rules for a world. I think all three types are equally valid worldbuilding.
So for all "is my idea feasible" questions I think meeting the overall rules for Worldbuilding.SE is enough, and they don't need to jump through additional hoops in my view.So for all "is my idea feasible" questions I think that meeting the overall rules for Worldbuilding.SE is enough, and that they shouldn't need to jump through additional hoops. Many science-based questions are feasibility tests, even if some are less overtly calling it so. Banning or restricting this category makes no sense to me.
Specific rules
You could try to be more specific as it comes to this category, in order to dissuade lazy questioners from pasting in their idea unformatted and expecting feedback on every part of it, from every angle. Particularly targeted therefore should be the need for narrow and objective scoring criteria.
Questions that ask only for feasibility without saying what makes an answer feasible, which system of rules to use; those questions don't specify any quality criteria for the answers. That renders them opinion-based. Figuring out new site policy to classify some of those as valid questions would be a challenge, and I'm personally not going to try to meet it.
Ichtys King's question specifies that it is asking for testing against "realistic biology and physics", so in my opinion it meets this threshold. I don't see the question as a particularly novel form compared to the existing body of questions.
So to respond to two of the bullet points:
A decision to support — or not support — questions of this type
- I support it if they have objective scoring criteria: feasibility is defined as a test against a set of rules. Those include but are not necessarily limited to branches of science, and the idea itself. (I personally think law should be included too)
Rules we expect OPs to follow when asking a question of this type.
- That the feasibility criteria are still reasonably narrow and alike in perspective. If you ask for idea validation from the perspective of both natural physics, economics and sociology; then you've most likely asked multiple questions at once.