Progressive concept teaching through three depth levels (Core → Mechanics → Deep Dive). Creates diagrams, provides annotated code walkthroughs from the current codebase, and builds explanations from fundamentals to production internals. Triggers: "teach me about [pattern/concept]", "how does [architecture/pattern] work", "walk me through [this implementation]", "tutorial on [concept]", "deep dive into [system/pattern]", "help me understand [this design]" Use when: user requests multi-level learning about code patterns, architecture, or implementation mechanics with checkpoint-based progression. Not for: quick answers, single-sentence explanations, code fixes, or "what does this line do" questions—those are standard assistance.
98
Quality
100%
Does it follow best practices?
Impact
97%
1.44xAverage score across 5 eval scenarios
Quick Reference
- Level 1 (Core): What & why → codebase example
- Level 2 (Mechanics): How it works → code walkthrough + tradeoffs
- Level 3 (Deep Dive): Production perspective → scaling, alternatives, pitfalls
- Checkpoint after each level: "Clear so far? Go deeper, or move on?"
Pause development mode. User seeks structured learning about code patterns, architecture, or implementation mechanics. Focus on practical understanding—pull examples from the current codebase, prioritize concepts that compound.
Most common: "explain", "teach me", "how does X work", "walk me through", "tutorial"
See trigger-reference.md for full trigger list and usage patterns.
Follow this sequence:
What it is, why it exists, and when to use it.
Example Format:
# Dependency Injection
**What**: Pass dependencies from outside instead of creating them inside.
**Why**: Makes code testable - swap real DB with mock in tests.
**When**: Any class that talks to external services (DB, API, filesystem).
Example from codebase:
├─ app/services/user_service.py:12
UserService(db_connection) ← injected, not createdCheckpoint: "Clear so far? Want mechanics, or is this enough?"
How it works under the hood.
Checkpoint: "Clear so far? Want the deep dive, or ready to move forward?"
Production behavior and senior perspective.
Checkpoint: "That's the full picture. Questions, or resume development?"
See examples.md for complete Level 1-3 walkthrough of dependency injection.
Visualization: Start with ASCII, upgrade to Mermaid/SVG only when necessary. See visualization-guide.md for decision tree.
Code snippets: Always from current codebase when possible - reference actual files/functions with file:line format.
Sample Checkpoint Dialogue:
Assistant: [Level 1 explanation of dependency injection]
Clear so far? Want mechanics, or is this enough?
User: "go deeper"
Assistant: [Level 2 explanation - annotated code walkthrough]
Clear so far? Want the deep dive, or ready to move forward?
User: "makes sense, lets move on"When the user signals understanding, immediately exit teaching mode and return to development:
Exit signals: "got it", "makes sense", "clear", "lets continue", "back to work", "/resume"
Response: Acknowledge briefly ("Great - back to coding") and resume the interrupted development task without re-explaining.
Install with Tessl CLI
npx tessl i neomatrix369/learning-opportunity