| name | notebooklm-learning-auditor |
| description | Specialized skill to audit and transform NotebookLM notebooks into "Continuous Learning Platforms". It focuses on pruning outdated content, configuring a "Learning Mentor" persona, and setting up recursive learning tools like Quizzes and Flashcards. |
| license | Apache-2.0 |
| metadata | {"version":"1.0","capabilities":["learning-optimization","persona-tailoring","knowledge-scaffolding"]} |
NotebookLM Learning Auditor Skill
This skill transforms traditional static notebooks into dynamic learning environments by applying structural audits, persona refinement, and knowledge retention strategies.
Operational Flow
Phase 1: Knowledge Mapping
- Source Discovery: Identify all sources and their "Freshness" (prefer 2025-2026 content).
- Gap Analysis: Identify missing core concepts needed for a "complete guide" (e.g., advanced ComfyUI nodes, latest SDXL updates).
Phase 2: Structural Pruning
- Noise Reduction: Delete sources that are too generic or redundant.
- Context Consolidation: Identify "Golden Sources" (primary documentation) vs "Support Sources" (tutorials).
Phase 3: Learning Persona Configuration
- Configure Chat: Set the custom instruction to a "Learning Mentor" profile.
- Tone: Socratic, encouraging, and technically precise.
- Method: Break complex concepts into "Atomic Units". Always end responses with a "Knowledge Check" question.
Phase 4: Retention Setup (NotebookLM Studio)
- Initialize Retention Tools:
- Create a Study Guide summarizing the entire notebook.
- Generate a Quiz (Medium difficulty) to baseline the user's knowledge.
- Creating Flashcards for key technical terms or workflows.
Verification
- Run a "Learning Path" query: "What are the 3 essential milestones to master [Topic] using these sources?"
- Verify that the AI provides citations and structured learning steps.
Principles
- Signal over Volume: A learning notebook is better with 15 expert sources than 100 random ones.
- Interactivity: Learning is active; the persona MUST challenge the user.
