Personal Learning Knowledge Graph Builder

Use this skill to design and evolve a structured, layered knowledge graph and generate engineer-grade knowledge node pages.

Knowledge Node Page Standard (engineer-grade)

Required section order (no deviations): 1) One-line definition 2) Problem setting 3) Mathematical formulation 4) Algorithmic interpretation 5) Optimization & implementation details 6) Connections & boundaries 7) Failure modes 8) Minimal pseudo-code 9) Decision checklist 10) Personal notes (leave TODO)

Mandatory constraints

• No vague explanations or generic motivation.
• Math derivations must state origin (e.g., MLE, MAP, constrained optimization).
• Explicitly state:
  • Problem boundary
  • Failure modes
  • Differences vs adjacent methods
• Writing target: algorithm engineers and researchers.
• Use LaTeX for math: $...$, $$...$$ only.
• No Unicode math symbols.
• Define symbols before use.

Required output sections

1) Knowledge Tree

• Output a Markdown tree.
• Annotate each level with a short learning goal.

2) Module Descriptions

• Provide descriptions for each first- and second-level module.
• Focus on engineering scope and cross-links.

3) Single Node Page Spec

• Provide the Knowledge Node Page Standard above.

4) Extension Rules

• Explain how to add new modules (e.g., Robotics/Control/LLM/VLM).
• Require cross-references and consistent naming.

Maintenance conventions

• Use consistent naming and symbols across nodes.
• Cross-link adjacent nodes explicitly (e.g., linear regression <-> ridge/lasso, MLE <-> MAP).
• Prefer concise, verifiable statements over pedagogy.