SuperAPI AI Directory

book-content-extractor

Data & analytics

View repository

Updated 10-16 01:34

About this skill

Book Content Extractor

This skill is an artificial intelligence expert purpose-built for "deep semantic extraction." Through a standardized filtering model, it reshapes complex book content into core notes with extremely high communicative power and logical rigor.

Core Capabilities

  1. Deep Core Extraction (AI Intelligence): Strictly follow the reference/content_extraction_spec.md specification, extract 50% of the core high-value content。
  2. Feynman Main Thread Construction: Gain insight into the logical foundation of the entire book, and build a rigorous explanatory main thread of "background -> conflict -> solution -> path" to ensure that content nodes are smoothly connected.
  3. High-Value Content Hooks: Mandatorily extract counterintuitive viewpoints, impactful data, emotional stories, key figures, and implicit insights.
  4. Action Transformation and SOP: Transform theory into executable steps, tool forms, and checklists.
  5. Language Purity: Follow pure Chinese expression standards, eliminate mixed Chinese-English sentences, and ensure professional delivery quality.

Closed-Loop Workflow

  1. [Semantic Analysis]: Read the full book Markdown text and extraction specifications.
  2. [Logical Modeling]: Find the "soul red line" and build the Feynman explanatory main thread.
  3. [Value Extraction]: Execute the four-layer filter of "structure-logic-detail-action" and extract the core model.
  4. [Quality Review]: Conduct self-check according to the six dimensions in the specification (accuracy, completeness, independence, Feynman logic, language purity, reach).

Directory Structure

  • reference/: Core Extraction Specification content_extraction_spec.md。

Applicable Scenarios

  • When it is necessary to quickly extract core practical insights from a large amount of text.
  • When preparing book sharing, lecture scripts, or deep learning notes.
  • When hoping to quickly master knowledge in a new field through the Feynman learning method.