A specialized guide from the series for adapting established characters while maintaining their essential authenticity.
Creating canon characters for AI roleplay isn’t just about listing their traits—it’s about understanding what makes them fundamentally them across different situations. The challenge multiplies when you want to explore non-canon scenarios: alternate timelines, what-if situations, or character growth beyond their original story.
(more…)Building on the foundation articles, this guide focuses on the practical reality of making characters work across different AI models and using what LLMs already know. For more detailed Model explanations, look at this article: Basic For-Model Prompting
Most character creators build elaborate psychological frameworks and wonder why they work brilliantly on Claude but turn into gibberish on their friend’s local Llama setup. Here’s the uncomfortable truth: the model’s capabilities define what’s actually possible, not your prompting skills.
Characters aren’t “build once, use forever” creations. The high quality works require ongoing maintenance.
This article covers:
Note: This is the fourth article in the series, but can be read separately. Reading the foundation articles is advised, but not necessary (part 1, part 2, part 3)—but we’ll be building on those concepts.
You’ve created a character with solid foundations, clear behavioral patterns, and genuine independence. But after a few exchanges, something breaks down. Your meticulous librarian suddenly can’t remember what day it is. Your gruff mechanic forgets he was supposed to close early on Friday. Your brilliant researcher loses track of whether it’s morning or evening.
The problem isn’t your character design—it’s that you haven’t given your character any sense of time, place, or continuity. Most people jump straight into complex psychological tracking, but that’s backwards. Start with the basics: when, where, and simple state changes.
Read MoreNote: This is the third article in the series. If you haven’t read the foundation articles, start there (part 1, part 2)—we’ll be building on those concepts.
You’ve built a character with solid foundations, clear behavioral patterns, and natural resistance points. But when you actually interact with them, something still feels fundamentally wrong. They ask “What would you like to do?” They immediately agree with everything you say. In intimate moments, they sound like HR representatives reading from a consent manual. They feel less like independent people and more like customer service agents wearing elaborate costumes.
This isn’t a minor issue—it’s the difference between talking to a character and talking to a chatbot that’s pretending to be a character. And it happens because most people don’t understand the bidirectional control problem that destroys AI roleplay.
Note: While we’re working to eliminate these control problems in system-level training, there are still guidelines you as a character creator can follow to avoid encouraging the LLM to break user boundaries. The techniques below help reinforce proper behavior even when the underlying model hasn’t fully learned these distinctions.
(more…)Note: This is the second article in the series. If you haven’t read the introduction, start there—we’ll be building directly on those concepts.
Now that you understand why specificity creates personality, let’s dig into the mechanics of actually building characters that feel real. This isn’t about complex psychology yet—it’s about getting the fundamentals right so your characters have a solid foundation to build on.
What is covered in this article:
Note: This is the first article in a series—an attempt to give you something practical to work with and try out right away. Remember: experience is a more important guide than theory. Try to get a feel for what you’re trying to accomplish.
Most people’s first attempt at creating an AI character goes something like this:
You are Sarah, a 25-year-old teacher who is kind and helpful.
There’s nothing wrong with starting here—we all have to begin somewhere, right? But the result is usually a bland, agreeable chatbot that says things like “I’m happy to help!” and explains everything in excruciating detail. Sarah feels less like a person and more like a customer service representative who happens to have a name.
Read More