StoryFlow

PROTOTYPE

Interactive story editing with inline widgets

How to use

Click on πŸ‘€characters or 🌍settings to edit properties. Watch how changes ripple through the story.

The fluorescent lights hummed overhead as stepped into . Her green eyes scanned the room cautiously. Something felt wrong. The usual sterile atmosphere was disrupted by an unfamiliar presence. burst through the door with his characteristic warm smile, breaking the tension. "You won't believe what I found!" he said optimistically, pulling out a weathered folder. "Files on Project Resonance." Elena's cautious nature made her hesitate. She needed to investigate further. The underground bunker felt more like a prison than a research facility.
πŸ‘₯

Characters

Elena
cautious
Marcus
optimistic
🌍

Settings

The Laboratory
sterile
⚑

Ripple Effects

Edit character traits or settings to see how changes ripple through the narrative automatically.

Active Ripples4
Story Elements21

How the Ripple Effect Algorithm Works

1

Change Detection

When you edit a character trait or setting property, the system identifies which story elements are connected to that change through predefined ripple effects.

2

AI Content Generation

For each affected story element, the AI analyzes the original text and generates updated content that reflects the new traits while maintaining narrative consistency.

3

Comprehensive Text Scanning

The AI scans the entire text for ALL references to old traits, including variations like "piercing green eyes" β†’ "piercing brown eyes", ensuring no references are missed.

4

Change Management

Generated changes are stored as "pending" and displayed with inline diffs. Users can accept or reject each change individually, giving full control over the narrative evolution.

5

Real-time Updates

Accepted changes immediately update the story text, while rejected changes revert to the original content. The system maintains consistency across all story elements.

6

Visual Feedback

The interface provides visual indicators during AI generation, highlights affected elements, and shows before/after comparisons with hover tooltips for easy review.

Technical Implementation

The algorithm uses OpenAI's GPT-4o-mini model with carefully crafted prompts that include the original text, old traits, and new traits. Quality control ensures changes are meaningful and maintain narrative flow. The system processes changes in parallel for efficiency.