I don't think this is really a homebrewing question at this point
If we're talking about the most recent edit, which asks about what published monster best matches the one from the novel, then it's not really a homebrewing question.
If it's a monster ID question then the content could be a lot more focused-- we really need mechanical details and not fluff, and ideally some guidance about what details are most important to match. For example, the monster having Disguise Self doesn't seem very important, since it could be added as a spell-like ability to any stat block without affecting CR.
Other details seem fluffy and likely won't translate to an existing monster well at all:
Rather than drain an single human dry, it seems to absorb a little bit of energy each from individuals in a large crowd
This sounds to me like a plot element-- this is how the creature survives and avoids detection as it feeds. If it's fluff, then it's not clear to me that this should have any mechanical representation at all; Life Drain doesn't work that way. If it's not fluff, then we're stuck deciding between finding something as close as possible (such as Life Drain), accepting that it might be possible that no published monster can have such a trait, or homebrew something with no particular criteria for doing so.
tl;dr: This may be a monster ID question, in which case it would benefit from additional edits which focus it in that direction. If this is the case, then the question is not ready to be reopened until such edits occur. It is not clear to me if this suits the intent of the asker.
Homebrewing provides tons of options, and the question isn't specific enough for one to be clearly better than another.
It's nice that the question has provided some details, including the target CR range. Unfortunately, these details are not enough to indicate what a "good" answer would be, or what would make one answer meeting the question's criteria better than another:
I would suggest a CR of at least 10, possibly closer to 15.
This is a broad range. There are enormous differences between a CR 10 creature and a CR 15, and there are a huge number of traits which can be changed to nudge the CR up or down. How could a potential answerer decide what CR to use? If someone homebrews this monster to a target CR of 14, is that better than one that has a target CR of 15? How about 13?
If one monster might offer a decent template that covers 75% of what the asker wants, and another offers 90% but would need some stat adjustments to fall into the right CR range, which is better? If we're going to be open to adjusting at least some stats, how many are available to be changed before it's "too much mechanical rebalancing"? If we settle on some number of changeable stats (say, 2), how can we judge between adjusting CR by increasing AC or HP versus adding or removing a resistance when the CR outcome is the same?
In the act of preforming what appear to be temple services, this being preformed magic that healed two people badly wounded and weakened by the attack of a vampire, though no specific divine spells as you would find in the Player's Handbook were mentioned by name.
Near the end of the story it also used a form of mind-influencing magic to ensure any humans in its 'flock' would not enter its temple quarters and possibly discover its true nature.
Finally, the monster in question seems to 'feed' on the energy (yes, I know you're going to complain how vague that sounds but it's as specific as I know how to say it) of its human followers in the act of preforming temple services. Rather than drain an single human dry, it seems to absorb a little bit of energy each from individuals in a large crowd, also it appears capable of feeding on a living being's energy without having to reach out and physically touch its victim (again, as specific as I know how to say it).
This is essentially all fluff. That's fine, but if something has no game-mechanical purpose then it's not relevant to the homebrewing process being asked about. And it again introduces lots of possibilities that we can't really judge between. If it is important to the asker that these traits do get some mechanical representation, we don't have enough information about what the traits are supposed to do or how they're supposed to work.
tl;dr: As a homebrewing question, it is underspecified. There is too little detail and too few clear goals and constraints for the question to fit the SE format. Additionally, it requires a lot of work on answerers' parts to offer up complete solutions so that the asker can choose one that "feels right" to them.
Either way, I think that the question is currently closed appropriately. Of course it could be reopened if enough elements of it changed, but that's always true. The operative questions are if those changes still get to what the asker wants, and if those changes are actually applied to the question.
