ljg-read is a specialized companion agent designed to transform passive reading into an active intellectual journey. Rather than simply summarizing content, it walks alongside the user through articles, essays, and academic papers, providing structural annotations and nuanced translations. It employs a unique 'Faithfulness-Expressiveness-Elegance' (Xin-Da-Ya) framework for English-to-Chinese translation and uses Socratic questioning to challenge the reader's assumptions. By classifying text segments into 'Skeleton', 'Muscle', and 'Fascia', the skill helps users visualize the architecture of an argument while documenting the entire learning process into structured Org-mode files for long-term knowledge retention.
主要功能
01Automated Org-mode Documentation: Generates comprehensive reading logs with time-stamped notes and understanding trajectories.
02Socratic Dialogue: Initiates 'intellectual collisions' by asking targeted questions to test the reader's understanding.
03Structural Mapping: Classifies text segments into logical units (Skeleton, Muscle, Fascia) for better navigation.
04Three-Layer Translation: Provides literal, natural, and deep-insight translations for complex English passages.
055,656 GitHub stars
06Multi-Source Support: Fetches and processes content from URLs, local files, and PDF documents.
使用场景
01Active learning sessions where the user wants to be challenged on their interpretation of an author's argument.
02Building a structured knowledge base of reading notes using Org-mode for research and study.
03Analyzing dense academic papers or philosophical texts that require structural breakdown and nuanced translation.