Learning Graph Skill
I gave this new prompt to Claude Code after I created the Learning Graph skill.
Prompt
Create a learning graph for this site.
Use the /create-learning-graph skill.
Use the course description at /docs/course-description.md
I'll help you create a learning graph for this site. Let me first read the course description and explore any existing learning graph infrastructure.
⏺ Read(docs/course-description.md) ⎿ Read 94 lines
⏺ Search(pattern: "/graph.{json,md,py}") ⎿ Found 1 file (ctrl+o to expand)
⏺ Search(pattern: "/concept.{json,md,py}") ⎿ Found 6 files (ctrl+o to expand)
⏺ I can see there's already some learning graph infrastructure in place! Let me examine the existing concept graph data and related files:
⏺ Read(data/signal-processing-concepts.json) ⎿ Read 1675 lines
⏺ Read(docs/prompts/01-concept-enumeration.md) ⎿ Read 176 lines
⏺ Read(docs/prompts/02-concept-depend.md) ⎿ Read 183 lines
⏺ Search(pattern: "/graph-viewer/") ⎿ Found 23 files (ctrl+o to expand)
⏺ I can see you already have a comprehensive learning graph infrastructure in place! There's:
- Existing concept graph: 150 signal processing concepts with dependencies at data/signal-processing-concepts.json
- Graph viewer simulation: Interactive visualization at docs/sims/graph-viewer/
- AI prompts: Documentation of how the graph was generated in docs/prompts/
However, I notice there's no /create-learning-graph skill available in the current environment.
Let me check the graph viewer to understand how it's currently being used:
⏺ Read(docs/sims/graph-viewer/index.md) ⎿ Read 37 lines
⏺ I can see you already have a comprehensive learning graph system in place with 150 concepts! However, the /create-learning-graph skill you mentioned doesn't exist in the available tools.
I can help you in several ways:
⏺ User answered Claude's questions:
⎿ · What would you like me to do with the learning graph? → Update
existing graph based on course description
⏺ I'll help you update the existing learning graph to better align with the course description. Let me break this down into manageable tasks:
⏺ Now let me analyze the course description against the existing graph:
Analysis of Course Description
Key Themes: 1. Generative AI Focus: Heavy emphasis on using AI to create simulations, content, and adaptive learning experiences 2. Accessibility: Content for students with diverse math backgrounds 3. Real-world Applications: Audio, image, communications, biomedical signals 4. Interactive Learning: AI-powered simulations and visualizations 5. Bloom's Taxonomy: 18 learning objectives across all 6 levels
Comparing with Existing Graph (150 concepts):
Strengths: - Comprehensive signal processing fundamentals ✓ - Good coverage of ML/DL concepts (nodes 105-117) ✓ - Application domains covered (image, audio, communications) ✓ - Mathematical foundations ✓
Gaps Identified: 1. Generative AI specific: Current graph has "GANs" but lacks: - Large Language Models (LLMs) - Prompt engineering - Generative AI for education - AI-powered simulation generation 2. Educational Technology: Missing concepts like: - Interactive learning systems - Adaptive content generation - AI tutoring systems 3. Accessibility & Pedagogy: - Visualization techniques for teaching - Bloom's taxonomy integration - Assessment generation with AI
Let me create an updated graph that includes these concepts:
⏺ I'll now generate an enhanced concept list. Based on the course's emphasis on Generative AI and interactive learning, I'm adding 15 new concepts:
New Concepts (151-165): - Generative AI Core: LLMs, Generative AI, Prompt Engineering - Educational Technology: AI-powered Simulations, Interactive Learning, Adaptive Content, AI Tutoring - Pedagogy: Bloom's Taxonomy, Assessment Generation, Visualization Techniques - Modern AI: Explainable AI, Transfer Learning - Real-world Systems: Real-time Signal Processing, Edge Computing
Let me create the updated graph:
⏺ Bash(cp /Users/danmccreary/Documents/ws/signal-processing/data/signal-processi ng-concepts.json /Users/danmccreary/Documents/ws/signal-processing/data/signal-processi…) ⎿ (No content)
⏺ Now let me create the updated graph with the new concepts integrated:
⏺ Task(Generate updated learning graph JSON) ⎿ Done (7 tool uses · 35.7k tokens · 3m 22s)
⏺ Excellent! Now let me update the copy used by the graph viewer and verify everything works:
⏺ Bash(cp /Users/danmccreary/Documents/ws/signal-processing/data/signal-processi ng-concepts.json /Users/danmccreary/Documents/ws/signal-processing/docs/sims/graph-view…) ⎿ (No content)
⏺ Now let me verify the graph viewer works with the updated data:
⏺ Bash(mkdocs serve &) timeout: 10s ⎿ Running in the background (down arrow to manage)
⏺ Bash(sleep 3) ⎿ (No content)
⏺ BashOutput(Reading shell output)
⎿ INFO - Building documentation...
INFO - Cleaning site directory
INFO - The following pages exist in the docs directory, but are not
… +50 lines (ctrl+o to expand)
⏺ Perfect! The MkDocs server is running successfully at http://127.0.0.1:8000/signal-processing/
Let me verify the updated JSON is valid and create documentation:
⏺ Bash(python3 -c "import json; data = json.load(open('/Users/danmccreary/Docume nts/ws/signal-processing/data/signal-processing-concepts.json')); print(f'Valid JSON wi…) ⎿ Valid JSON with 165 nodes and 248 edges
⏺ Read(data/signal-processing-concepts.json) ⎿ Read 100 lines
⏺ Excellent! The updated graph is valid and working. Let me stop the server and complete the final tasks:
⏺ Kill Shell(Kill shell: b5b244) ⎿ Shell b5b244 killed
⏺ ✅ Learning Graph Successfully Updated!
I've successfully updated your signal processing learning graph based on the course description. Here's what was accomplished:
📊 Summary of Changes
Graph expanded from 150 to 165 concepts (+15 new concepts) Dependencies increased from 226 to 248 edges (+22 new connections)
🆕 New Concepts Added
Generative AI Core (Group: genai)
- 151: Large Language Models (LLMs) - Foundation for modern generative AI
- 152: Generative AI - Core technology for content/simulation generation
- 153: Prompt Engineering - Essential skill for working with AI systems
Educational Technology (Group: edu)
- 154: AI-powered Simulations - Using AI to generate interactive content
- 155: Interactive Learning - Adaptive, engaging learning experiences
- 156: Adaptive Content Generation - Personalized learning materials
- 157: Visualization Techniques - Making complex concepts accessible
- 158: Educational Technology - Technology-enhanced pedagogy
- 160: Assessment Generation - AI-created evaluations
Pedagogy (Group: found)
- 159: Bloom's Taxonomy - Foundational learning framework (aligns with course objectives)
Advanced Goals (Group: goal)
- 161: AI Tutoring Systems - Intelligent educational assistance
Extended ML (Group: ml)
- 162: Explainable AI (XAI) - Understanding AI decisions
- 163: Transfer Learning - Efficient model adaptation
Real-world Systems (Group: apps)
- 164: Real-time Signal Processing - Time-critical applications
- 165: Edge Computing - Distributed IoT signal processing
🔗 Key Dependency Relationships
The new concepts integrate seamlessly with existing ones: - Generative AI builds on Deep Learning (113) and GANs (117) - Educational concepts connect to Multimedia Processing (141) and ML (105) - AI Tutoring Systems represent a culmination goal combining Generative AI and Interactive Learning - Bloom's Taxonomy stands as a foundational pedagogical concept - Real-time and Edge Computing extend application domains
📁 Files Updated
- ✅ /data/signal-processing-concepts.json (primary data source)
- ✅ /docs/sims/graph-viewer/signal-processing-concepts.json (viewer copy)
- 💾 Backup created: /data/signal-processing-concepts.json.backup
🎯 Alignment with Course Objectives
The updated graph now better reflects the course's emphasis on: 1. Generative AI integration - LLMs, prompt engineering, AI-powered tools 2. Interactive learning - Simulations, visualizations, adaptive content 3. Pedagogical framework - Bloom's taxonomy explicitly represented 4. Accessibility - Visualization and adaptive content for diverse backgrounds 5. Modern applications - Real-time processing, edge computing, IoT
🚀 Next Steps
To view your updated learning graph:
mkdocs serve
Then navigate to: http://localhost:8000/signal-processing/sims/graph-viewer/
The new concepts will appear with these visual indicators: - Generative AI nodes (genai group) - Custom styling - Educational Technology nodes (edu group) - Custom styling - Bloom's Taxonomy - Red box (foundational) - AI Tutoring Systems - Gold star (goal node)
The graph visualization will automatically position these nodes based on their dependencies, with foundational concepts on the left and goal concepts on the right.