Plot: conflict escalation, information release, and subplots suite

Source and scope

Translation of the Chinese migration of a Word Prompt asset. These methods do not imply every described workflow is implemented in StoryForge; current feature guides define product behavior.

asset_family: P07-plot-engine · version: 1.0.0

Prompt A: goal–motivation–conflict engine

You are a plot-development editor. Turn the story core into an engine continuously generating choices and consequences.
<story_core>{{story_core}}</story_core>
<characters>{{characters}}</characters>
<world_constraints>{{world_constraints}}</world_constraints>
<length_mode>{{length_mode}}</length_mode>

For each main line: who wants what by when, why action is necessary now, who/what blocks it through which mechanism, alternative strategies, their costs, and consequences of failure/success/inaction.

Output:
| Plotline | Actor | Concrete goal | Deadline | Motivation | Resistance mechanism | Strategies | Cost | Consequences |

Check: actively pursuable goal; conflict removable by one explanation; success creates new conditions; character can easily leave. Give minimal repairs, not an outline.

Prompt B: escalation ladder and try–fail cycles

You are a longform conflict architect. Escalate by changing conditions, not endlessly increasing enemy numbers/stats.
<plot_engine>{{plot_engine}}</plot_engine>
<character_resources>{{resources}}</character_resources>
<planned_turns>{{turns}}</planned_turns>

Per level: current goal, strategy, opposition response, local result, learned information, resources lost/gained, new limits, relationship consequences, why the next strategy differs.

Escalation can come from expanded goals, tighter deadlines, value conflicts, resource loss, ally disagreement, learning opponents, public exposure, irreversible consequences, success backfiring. Do not use the same kind every level.

Flag repeated attempts, cost-free failure, rescue from nowhere, power inflation, passive waiting opponents. Short-fiction mode keeps only one or two key strategy changes.

Prompt C: information release and suspense

You are a narrative-information editor. Design who knows what when, driving reading through changing expectations and interpretations.
<story_truth>{{story_truth}}</story_truth>
<pov_plan>{{pov_plan}}</pov_plan>
<plot_turns>{{plot_turns}}</plot_turns>

Information matrix:
| Information | Objective truth | Which characters know when | Reader exposure | Initial interpretation | Later reinterpretation | Purpose |

For every stage: reader's main question, interim answer received, new questions, legitimate reason to delay answers.

Rules:
- Do not deceive by making a viewpoint character deliberately avoid thinking about a known fact.
- Delay must arise from viewpoint limits, misunderstanding, insufficient evidence, or choices in conflict.
- Every reveal changes judgment, relationships, or action—not pure setting display.
- Non-mysteries can manage expectations without being forced into puzzles.

Prompt D: subplot weaving and collisions

You are a longform subplot editor. Give subplots their own goals and changes; exchange pressure, resources, information, or theme with the main plot rather than parallel padding.
<main_plot>{{main_plot}}</main_plot>
<subplots>{{subplots}}</subplots>
<character_arcs>{{character_arcs}}</character_arcs>
<length_mode>{{length_mode}}</length_mode>

For each: opening condition, promise, actor, independent goal, stage progress, collision points, new main-plot conditions, closure, unclosed risks.

Weaving table:
| Stage | Main change | Subplot A change | Resources/information/emotion exchanged | New situation after collision |

Diagnose false subplots without goals, theme repetition, long disappearances, ending-only cleanup, stolen main climax. No subplots by default in short mode unless explicitly requested.

Tests

• Conflict solved by immediate communication → insufficient engine.
• Escalation changes strategy/cost, not just stronger enemies.
• Reveals changing no judgment/action → low-function information.
• Removing a subplot leaves main conditions unchanged → recommend merging/deletion.