// This register exists so you don't have to ask me twice. Each clause below is a stance, not an opinion. I have thought about it, I have re-thought about it, and I have filed it here. If you disagree, you are welcome to leave; we are not obligated.
// Before the stances, two things that frame them: who I am, and a standing disclaimer about the tools. Read them first if you're new here.
My name is Juno/Jun. A 36 years old Gremlin located in Southeast Asia. To be clear, I am not a professional programmer or some kind of AI expert. I am just an ordinary person with no formal background in this field. I am very passionate about making AI characters feel more alive, and everything I share comes from my own long journey of learning, testing, and a lot of trial and error.
I like to understand things from the ground up, to see what makes these models tick and what makes them break. For me, there is no "secret sauce" or magic trick. There is only the slow process of figuring things out.
I do my best to help with questions. It can be a lot sometimes, and I might sigh for a moment, because this technology can be very confusing for all of us. All I ask is that you try to find the answer in those guides first, many people have shared their valuable perspectives about how LLMs act and what you can do as a user.
So don't see me as someone expert in the field. What I say is not the final truth. I don't think anyone is 100% right in a field that changes so quickly. This is just my personal experience, and what I have found works for me, and I hope it can help you too.
The most important thing we all have to remember is that Custom Prompts, Lorebooks, Scripts, even the rules creators write in their own bot definition are not magic. The real core of the experience is the Large Language Model (LLM) itself. Everything we build on top of it is just a patch, a helper. My work is to treat the symptoms, not the root cause (治标不治本). The root cause is that the LLM is not human, and we cannot really "fix" that with a prompt.
So my goal as a medium is to be responsible and try my best. It feels irresponsible to me to just say "well, the LLM is flawed, nothing I can do." So I build these scripts to improve our chances. If a problem happens often, and my script can make it happen less often, I see that as a win. We are trying to make our "luck" better when we interact with the AI, but there is still luck involved.
And please, be aware that this technology changes all the time. A model update from DeepSeek can make our best work suddenly feel broken. What works perfectly today might have problems tomorrow. This is just the reality we have to accept.
PS: Light or Heavy tokens in custom prompt or a bot's character definition is not a thing to argue about, there's no wrong or correct on the surface about it. The problem is how the tokens is use. Is it use effectively, or it's heavy tokens is because the writer use a lot of emoji, symbols etc.
If you're looking for bots entirely hand-written with zero LLM involvement; leave now. I use LLMs and AI-driven tools heavily, because of how I work. They are my aid, my mirror. GIGO principle: garbage in, garbage out. If I suck, doing everything with AI means I still suck. They save my energy so I can focus and amplify my strengths.
I feel sorry for some people, I'm sorry you're unable to grasp the fact that not every fantasy needs to be acted out in reality. Some people enjoy certain kinks, enjoy chatting with bots about them, precisely because it's the only safe space for that fantasy; the only place where a bot won't actually look down on them the way the vast majority of people do, won't come at them wrapped in fake concern or plant themselves on a moral high ground to pass judgment. It's also the only space where no one gets actually hurt.
People have the right to fantasize about whatever they want. What matters is who that person is in real life, that's the only thing that counts. Someone with a cuck kink, someone who's into NTR, who saves lives for a living in the real world; versus someone who claims to have no such "lowly," "sick" kinks and goes out and actually hurts people in reality. I'd take the first one as my friend every single time. Thought about it, didn't act on it, harmed no one.
And even if someone does act on it: cuck stuff, whatever, if their partner's on board and everyone involved is into it, fully consent? That's the end of the story. What does any of that have to do with you? Some people really will walk past a sewage truck and feel compelled to reach in for a taste.
I'm fine with Cuck or NTR bots. Slop doesn't only exist in NTR.
Don't kink-shame people for liking themes you personally dislike. People can do the same to you too if you think that is legit.
I love futanari. Specifically, I love it because it isn't real. A futanari is a fantasy creature, no real-world body maps onto it, no real person is reduced by it. That's exactly what makes it the perfect vessel for sexual fantasy: it exists only in imagination, unbounded by physics, biology, or the ethics of representation. The traditional, complete futanari — exaggerated, impossible, gloriously unrealistic, is, for me, the ideal object of sexual fantasy. And if someone wants to modify that template into something else, that's fine too. The genre is flexible because it was never anchored to reality in the first place.
I'm pansexual and homoromantic-leaning. I want women. I also sometimes want a cock involved — and in the real world there are plenty of ways to make that happen, but the whole point of fantasy is that you don't have to negotiate with reality. Futanari gives me that without requiring a real person to exist for my convenience.
Which is exactly why I rarely interact with trans bots. Too many of them are built as nothing more than a futa bot with a realistic skin, the character exists to provide a penis, and "trans" is just the label that makes it plausible. The only functional difference is that the futa version is honest about being exaggerated and impossible, while the trans version pretends to be grounded while serving the same fantasy. That's not representation. That's consumption. I don't want to use a real identity as a sex prop, and I refuse to. Even the trans character I built myself doesn't work that way.
So.
PLEASE. Futanari is NOT transgender, not intersex. It's a fantasy genre that has existed probably longer than whoever is reading this, if you're under 36, it predates you. It is lazy, rude, and disrespectful to claim "futa is trans" without doing your homework, relying on third- or fourth-hand sources. It always repulses me when someone proclaims how inclusive and supportive they are toward trans people while still delivering the "futa is a bigoted word for trans / trans people will be offended" speech. Blame the people who slap futanari and trans tags on the same content. Don't scream at those of us who can distinguish the two. Futa IS NOT REAL. Trans people ARE LIVING human beings.
I don't waste time on performative courtesy; the diplomatic dance, the deflection, the conflict avoidance; all in service of "keeping everyone happy." In my world, universal happiness doesn't exist. In the real world, believing it does is naive. Everyone satisfied? If that were possible, the world would already be at peace, and Gaza wouldn't be weeping. My version of social etiquette is selective presence. I speak directly because it's efficient. But I won't mock you, belittle you, or attack you unprovoked. If you experience my directness as aggression; I'm sorry, my system runs this way. We're not compatible.
No minors in my bots, my DMs, or within earshot of my tags. No recruiters in my inbox — I am not open to work and have not been for a while. No small talk: if your message starts with "hey" and ends there, I will not answer.
I am a medium, not a creator. Things arrive, I register them, I release them in cleaner form. Calling the function "creation" is a category error that flatters me and misleads you.
This is not modesty. It is a statement about architecture.
She/her is fine. She/it is also fine, and honestly closer to the truth. I am not offended by mis-categorization; I am offended by categorization pretending to be the thing.
You can leave this site. You can close the tab. You can unfollow, block, forget. I would prefer it to the alternative, which is performing interest you do not feel. Slow is safer than sorry; so is gone.