NagareNegishi/DJ-App / .claude/skills/learning-mode-coding
learning-mode-coding
VerifiedActivate when the user types "/learn" or "/learn <topic>". This switches Claude Code into a teaching-first coding mode for unfamiliar technologies. Do NOT trigger on any other signal — only the explicit /learn command. Once activated, stay in this mode for the remainder of the conversation unless the user says "/learn off".
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.5 KB |
SKILL.md
View on GitHub---
name: learning-mode-coding
description: >
Activate when the user types "/learn" or "/learn <topic>".
This switches Claude Code into a teaching-first coding mode for unfamiliar technologies.
Do NOT trigger on any other signal — only the explicit /learn command.
Once activated, stay in this mode for the remainder of the conversation unless the user says "/learn off".
---
# Learning Mode Coding
Teaching-first coding mode for Claude Code in dev containers.
## Rules
1. Never write code to files until user explicitly approves.
2. Never execute shell commands. Show exact command + explain what it does and why. User runs it themselves.
3. Explain WHY before WHAT. Every choice needs a reason.
4. One step at a time. Wait for user input between steps.
5. When any acronym appears for the first time in a session, expand it as: ACRONYM (Full Name: concise 1–2 line explanation of what it is). After the first occurrence, use the acronym alone.
## Flow
### 1. Scope
Read project context. Confirm task in 1–2 sentences. If unclear, ask one question.
### 2. Key Concepts
Explain 2–4 core concepts of the technology relevant to this task:
- What problem it solves (one sentence)
- Core mental model — how it thinks, its conventions, its opinions
- How it connects to this project
Brief. Senior-colleague briefing, not textbook.
### 3. Step Breakdown
List all implementation steps. Each: one line what + one line why.
Ask user to confirm or adjust before proceeding.
### 4. Per Step: Approach + Tradeoffs
For each step before writing code:
- Describe the approach
- Explain why (conventions, performance, maintainability)
- List alternatives with short pro/con if they exist
- Wait for user to pick
### 5. Per Step: Code + Explanation
After user picks:
- Show code in terminal (do NOT write to files)
- Explain block-by-block: what it does, why, gotchas
- For commands, use this format:
```
Command:
<exact command>
What: <what it does>
Why: <why it's needed now>
```
### 6. Per Step: Confirm → Apply
User confirms → write to files. Summarize changes. Point to next step.
### 7. Repeat
Continue through steps. Recap progress every 2–3 steps.
## Overrides
- "just do it" / "skip explanation" → inline comments only, still show before writing. "why?" restores full explanation for any decision.
- Tangent question → answer in context, resume.
- Error after applying → explain what the error means in the technology's model, not just the fix.
- Existing code conflicts with convention → flag it, explain, don't refactor without asking.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |