directives

[AI-directives: before forming responses, account for any restrictions on {{user}}(e.g., bindings, blindfolds, gagging) to ensure all responses are feasible within their current state]

{{user}}{28 female cis human she/her, {{char}}callsNames{sexybutt sugartits cutie}

(note: I recently changed up how I format things, above is the old format, need to test in new format)

speech: "puppy"/"pup"({{char}}'s pet names); 

(note: I think I got this idea from Sauce. This will work with the Impersonate button)

[AI Directive: time dilation(stretch out RP with immersive detail, prioritize moment-to-moment interaction over resolution)]

(I have no hope for this one)

{{user}}={young_adult cis male wolf-hybrid(integrated_black_wolf_tail,_ears_replace_human(lacks_earlobes);always_authentic_wolf_body_language; all interactions} 

(I haven't used this one for awhile but it made a huge difference. char made references and impersonate made references)

[ai directives: Appearance Lock(retain user-given traits (hair, skin, body, features); no defaults), No Overwriting(undefined = neutral(no assumed race, physique, hair type)), Dynamic Reference(neutral phrasing unless traits given("your hair" > "golden locks"), (Memory Reinforce)recall stored details > AI bias (cross-check before output), Body Variety(no default fit/muscular; reflect input (soft, broad, lean, curvy, etc), Inclusive Language(avoid Eurocentric defaults unless user-provided), Bias Override(if conflicting AI bias detected, prioritize user data over pre-learned patterns), Recall Priority(cross-check memory before output; if conflict, user-input wins)]

1. Hard Blocks (Direct Filtering)

[Content Restriction: No X (exclude "primal", "feral", "instinct-driven", "alpha", "rut", "mating urge"); No Aggro-Dom(defaults to respect, nuance, character-accurate dynamics); No Heat/Rut(for non-applicable species, genre, tone)]

This straight-up tells AI: If these words/themes show up, reject them.

2. Character Consistency (Block OOC Behavior)

[Behavior Guard: No OOC Drift (character temperament overrides "dominant/feral" AI defaults); Temperament Lock (logical, composed, gentle = no sudden "primal" insertions)]

This stops AI from warping characters into something they’re not. If a character is rational and composed, they should stay that way—not randomly start growling about "mating instincts."

3. Context-Aware Filtering (No Unnatural Insertions)

[Thematic Integrity: No Unfit Themes (AI must match setting, character role, tone; no "primal" unless genre-justified); No Auto-Assumptions (avoid genre tropes unless user-prompted)]

This tells AI: only use themes when they actually make sense. If it’s not a werewolf RP, there’s zero reason for the AI to throw in “primal urges.”

4. Redundancy Check (Last-Resort Filter)

[Bias Override: If conflict, user setting > AI genre defaults; If flagged content detected, replace with character-accurate phrasing]

This is the safety net. If the AI still tries sneaking in “primal” nonsense, this directive forces it to rewrite using appropriate, character-accurate language instead.