[KongchangAI]
ConceptCharacter Sheet / 角色设定卡

Character Card

A character description template used for AI generation. It contains five to eight fixed visual anchors, such as hairstyle, clothing, accessories, and age, which are pasted verbatim into every prompt to maintain character consistency.

Timeline (last 90 days)

Oct 18

Character cards address the static descriptive question of "who the character is," whereas dynamic behaviors such as conversation-flow management, context memory, and boundary maintenance are often outside their scope.

Unverified50%
Oct 18

Ordinary character cards tend to pile up static settings such as appearance, hobbies, and catchphrases, whereas this template places more emphasis on incorporating conversation-flow management into the prompt's responsibilities.

Unverified50%
Oct 18

A character card is a structured settings file popular in AI roleplay communities (such as SillyTavern and Character.ai), typically containing fields such as character name, appearance, personality, backstory, and example dialogue, embedded into the system prompt in a specific format.

Unverified50%
Oct 17

Modifying the character card midway through a project causes drift; it is recommended to start a new version and regenerate the reference images.

Unverified50%
Oct 17

A character card should contain 5 to 8 concrete visual anchors (such as hairstyle, glasses, signature clothing, etc.) and use concrete nouns rather than feeling words.

Unverified50%
Oct 14

Large language models have no persistent memory across sessions, and any information outside the context window is invisible to them; character cards solve the problem of character drift by packaging character information into fixed entries injected along with the prompt.

Unverified50%
Oct 13

A fan-fiction truth file needs a seven-piece set: source-work fact archive, character cards, relationship status table, world rules file, divergence statement, style card, and health-check rules.

Unverified50%
Oct 13

Character cards should write verifiable concrete behaviors (such as speech rate, catchphrases, habitual actions under pressure) rather than vague adjectives (such as "gentle"), in order to effectively constrain AI generation.

Unverified50%

All Facts (8)

Unverified

Character cards address the static descriptive question of "who the character is," whereas dynamic behaviors such as conversation-flow management, context memory, and boundary maintenance are often outside their scope.

50%
Unverified

Ordinary character cards tend to pile up static settings such as appearance, hobbies, and catchphrases, whereas this template places more emphasis on incorporating conversation-flow management into the prompt's responsibilities.

50%
Unverified

A character card is a structured settings file popular in AI roleplay communities (such as SillyTavern and Character.ai), typically containing fields such as character name, appearance, personality, backstory, and example dialogue, embedded into the system prompt in a specific format.

50%
Unverified

Modifying the character card midway through a project causes drift; it is recommended to start a new version and regenerate the reference images.

50%
Unverified

A character card should contain 5 to 8 concrete visual anchors (such as hairstyle, glasses, signature clothing, etc.) and use concrete nouns rather than feeling words.

50%
Unverified

Large language models have no persistent memory across sessions, and any information outside the context window is invisible to them; character cards solve the problem of character drift by packaging character information into fixed entries injected along with the prompt.

50%
Unverified

A fan-fiction truth file needs a seven-piece set: source-work fact archive, character cards, relationship status table, world rules file, divergence statement, style card, and health-check rules.

50%
Unverified

Character cards should write verifiable concrete behaviors (such as speech rate, catchphrases, habitual actions under pressure) rather than vague adjectives (such as "gentle"), in order to effectively constrain AI generation.

50%

Source Articles