# OnlyKin AI Roleplay Glossary

URL: https://onlykin.ai/glossary
Updated: 2026-06-04

Concise definitions for AI character chat, roleplay memory, character cards, personas, lorebooks, context windows, and related creator terms.

## Terms

- [AI character chat](https://onlykin.ai/llms/glossary/ai-character-chat): AI character chat is a conversation experience where a language model responds as a defined character instead of a generic assistant.
- [Character card](https://onlykin.ai/llms/glossary/character-card): A character card is a structured bundle of character information that tells an AI roleplay model who the character is and how the first scene should begin.
- [Context window](https://onlykin.ai/llms/glossary/context-window): A context window is the maximum amount of text a language model can consider at once when generating a reply.
- [Example dialogue](https://onlykin.ai/llms/glossary/example-dialogue): Example dialogue is sample conversation text included in a character setup to teach the model how the character speaks, reacts, and formats replies.
- [Lorebook](https://onlykin.ai/llms/glossary/lorebook): A lorebook is a set of background entries that can be added to an AI roleplay prompt when relevant keywords, rules, or retrieval settings match the current scene.
- [Persona](https://onlykin.ai/llms/glossary/persona): A persona is the identity the user brings into an AI character chat, such as a name, role, traits, appearance, or relationship to the character.
- [Private character](https://onlykin.ai/llms/glossary/private-character): A private character is an AI character card that stays outside public discovery while the creator drafts, tests, or uses it personally.
- [Semantic memory](https://onlykin.ai/llms/glossary/semantic-memory): Semantic memory is compact stored information about important facts, preferences, or events that can be retrieved later because of meaning rather than raw message order.
- [System prompt](https://onlykin.ai/llms/glossary/system-prompt): A system prompt is high-priority instruction text that guides how an AI model should behave before it reads the user's message.
- [Token](https://onlykin.ai/llms/glossary/token): A token is a small piece of text, such as a word fragment, word, punctuation mark, or space pattern, that a language model processes.
- [World info](https://onlykin.ai/llms/glossary/world-info): World info is a roleplay memory system that stores background facts and inserts matching entries into the prompt when the current chat context makes them relevant.