Scripting Example: Text Evolution Pipeline

A multi-stage text evolution system that uses SPOUT's ⬠Δ Describe(Reduce) module for generating descriptions, ⬠Μ Mutate module for generating variations, and ⬠Ε Evaluate module for selecting optimal results. The pipeline progressively evolves text while maintaining semantic consistency and improving quality.

Evolution Pipeline Flowchart

Process Overview

  1. Initial Input
    • Takes input text from user prompt or arguments
    • Generates description using ⬠Δ Describe(Reduce) module
    • Prepares text for processing
  2. Text Segmentation
    • Splits input text into individual sentences
    • Preserves sentence structure and relationships
  3. Variant Creation (⬠Μ Mutate)
    • Generates multiple variations of each sentence
    • Maintains core meaning and intent
    • Explores alternative expressions
  4. Quality Selection (⬠Ε Evaluate)
    • Assesses all variants against criteria and description
    • Selects optimal version for each sentence
    • Maintains overall text cohesion
  5. Progressive Saving
    • Writes content to markdown file as processing occurs
    • Maintains detailed evolution history
    • Preserves intermediate results
  6. Final Composition
    • Combines best variants into final text
    • Maintains coherence with original context
    • Provides complete evolution record

SPOUT Modules Used

  • ⬠Δ Reduce: Generates descriptive context for the input text to guide evaluation
  • ⬠Μ Mutate: Creates variations of text segments while maintaining core meaning and intent
  • ⬠Ε Evaluate: Assesses variants and selects optimal versions based on specified criteria and context

Configuration Options

{
  input: string,        // Text to evolve (default: user will be prompted)
  criteria: string,     // Evaluation criteria (default: "quality, interesting, creative, coherent")
  variants: number,     // Variants per sentence (default: 5)
  mutation: number,     // Mutation level (default: 1)
  description: string   // Optional custom description
}

Usage Examples

Basic Evolution

./text_evolve_pipeline.mjs --input "The cat sat on the mat."
  • Uses default criteria and settings
  • Generates variants with standard mutation

Custom Criteria

./text_evolve_pipeline.mjs --criteria "quality,creativity" --variants 3
  • Specifies custom evaluation criteria
  • Adjusts number of variants

Advanced Evolution

./text_evolve_pipeline.mjs --input "Long text..." --mutation 2
  • Increases mutation level for more variation
  • Processes longer input text

Complex Configuration

./text_evolve_pipeline.mjs --input "Some text" --criteria "humor,engagement" --variants 8 --mutation 3
  • Multiple custom settings
  • Maximum variation and evaluation

Output Format

  • Files created in ../output/text_evolve_pipeline_output/
  • Named with timestamp: evolution_YYYYMMDDHHMMSS.md
  • Markdown format with sections:
    • Original text
    • Generated description
    • Sentence-by-sentence evolution
    • Final evolved text

Dependencies

  • SPOUT CLI: Required for text processing modules
  • zx: Used for shell operations and file handling
  • Node.js: Runtime environment

Error Handling

  • Sanitizes input text to prevent command errors
  • Preserves progress through progressive saving
  • Provides detailed error logging
  • Ensures partial results are saved

Use Cases

  1. Text Improvement
    • Content enhancement
    • Style refinement
    • Coherence maintenance
  2. Creative Writing
    • Story evolution
    • Sentence variation
    • Style experimentation
  3. Content Optimization
    • Message clarity
    • Engagement improvement
    • Theme consistency
  4. Educational Tools
    • Writing assistance
    • Style analysis
    • Text transformation examples
Table of Contents