data-ingestion-pipeline

What problem does it solve?

This Skill automates the complex process of ingesting educational math content from raw textbook markdown into a structured, queryable format within a data service layer, enabling personalized learning experiences.

Core Features & Use Cases

• End-to-End Content Pipeline: Manages the entire workflow from reading textbook markdown to generating and writing concepts, personalized content, exercises, and Manim animations.
• LLM as Content Generator: Leverages Claude Code as the primary AI to perform all content creation tasks, ensuring consistency and reducing external API calls.
• Data Service Integration: Writes all generated content directly to a data service layer (PostgreSQL and Neo4j) using TutorDataServiceWriter.
• Personalization Engine: Rewrites content and generates exercises tailored to specific student interests and grade levels.
• Animation Code Generation: Creates Manim Python code for visual math explanations.
• Use Case: Ingesting a new Algebra 1 chapter, extracting its core concepts, generating personalized explanations and practice problems for different student interests (e.g., basketball, gaming), and creating corresponding Manim animations for each section.

Quick Start

Use the data-ingestion-pipeline skill to process the textbook chapter located at 'curriculum/algebra1/textbooks/chapter_07_factoring.md'.