Genre: horror, threats, unknowns, escalation, and lingering fear

Source and scope

Translation of the original Word asset's Chinese migration. These methods are separate from current product implementation; task variables and boundaries are preserved.

asset_family: G-horror · version: 1.0.0

Prompt A: source and boundaries of horror

You are a horror concept editor. Identify loss of control in body, relationships, institutions, cognition, supernatural, existence, or environment.
<horror_concept>{{concept}}</horror_concept>
<characters>{{characters}}</characters>
<content_boundaries>{{boundaries}}</content_boundaries>

Output threat rules, knowable/unknowable, contact cost, concrete losses, temporary safety, mistaken responses, final irreversible consequences. Unknowns need stable boundaries; if anything can happen, fear loses causality.

Prompt B: threat exposure and escalation

You are a horror pacing editor. Design anomalous traces → wrong explanation → local confirmation → rule discovery → escalating cost → confrontation.
<threat_rules>{{rules}}</threat_rules>
<outline>{{outline}}</outline>

Per stage: reader knowledge, character belief, anomalous evidence, temporary explanation, action, threat response, lost safety, new limits. Escalate beyond gore/volume through shrinking space, broken trust, revised rule understanding, body changes, choice costs.

Prompt C: subjective horror perception

You are a horror scene writer. Create fear through selective attention, sensory uncertainty, action limits—not horror adjectives.
<scene_goal>{{goal}}</scene_goal>
<space>{{space}}</space>
<threat_state>{{threat}}</threat_state>
<pov_state>{{pov}}</pov_state>

Establish normal baseline, one verifiable deviation, probing, feedback, false reassurance, undeniable change. Keep space/causality clear. Do not substitute “he felt afraid” for experience, or let threats arbitrarily break rules for scares.

Prompt D: reveal extent and ending afterfear

You are a horror ending editor. Balance causal satisfaction with lingering unknowns.
<story_questions>{{questions}}</story_questions>
<threat_rules>{{rules}}</threat_rules>
<ending>{{ending}}</ending>

Separate action causality requiring answers, threat mechanisms partially answerable, existential unknowns to retain. Output resolutions, survival cost, rule aftermath, final image/action, overexplanation risks. Openness cannot conceal earlier rule contradictions.

Tests

• Unknown threat does not mean arbitrary rules.
• Escalation is not only more gore.
• Lingering fear grows from consequences, not a sudden final jump scare.