Menu
Guide users through progressive lessons to learn AI assistant basics
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Load context from previous sessions so you can pick up where you left off
Save what you worked on so your AI can pick it up next time
Example skill demonstrating the skill structure. Use as a template for creating your own skills.
Extract actionable content and filter hype/marketing from videos and articles. Invoke when user provides content with potential hype/marketing, wants signal without noise, or needs bullshit filtering.
Improve LLM prompts using prompt engineering best practices. Invoke when user wants to optimize prompts, improve slash commands, or apply prompt engineering.
Summarize meeting transcripts or notes into actionable format. Invoke when user provides meeting transcripts, notes, or wants actionable meeting summaries.
| name | lessons |
| description | Guide users through progressive lessons to learn AI assistant basics |
| allowed-tools | ["Read","Write","Bash"] |
Progressive lessons that teach users how to work with AI assistants by actually using the system.
Enable users to learn AI augmentation through hands-on experience, not just reading documentation. Lessons build on each other and teach core concepts through practical exercises.
Automatic invocation:
Progressive invocation:
Lessons are stored in skills/lessons/content/ and numbered for progressive learning:
Check what the user wants:
Check for progress tracking file:
PROGRESS_FILE="${SOURDOUGH_DATA:-$HOME/ai-data}/learning/lesson_progress.yaml"
If file exists, read it to know:
Read the appropriate lesson file from skills/lessons/content/NN-name.md
Present the lesson content to the user, following the structure in the lesson file.
Lessons are interactive experiences with pause points where you wait for user action:
Structure:
[PAUSE] marker[PAUSE]During pauses:
Tone:
At the end of each lesson:
After lesson completion, update progress file:
last_updated: YYYY-MM-DD HH:MM
completed_lessons:
- 01-first-conversation
- 02-reading-files
current_lesson: 03-creating-files
notes:
- "User particularly interested in file operations"
- "Completed lesson 2 quickly, comfortable with reading"
Offer:
User's progress is stored in their personal data layer:
~/ai-data/learning/lesson_progress.yaml%USERPROFILE%\ai-data\learning\lesson_progress.yamlCreate the file on first use if it doesn't exist.
# Lesson N: [Title]
[Friendly introduction to the lesson topic]
## What You'll Learn
- Key concept 1
- Key concept 2
- Key concept 3
## Let's Try It
[Interactive demonstration or exercise]
## Key Takeaways
- Important point 1
- Important point 2
## Questions?
[Pause for user questions]
---
Ready for the next lesson? Or would you like to practice this more?
# Sourdough Lessons
Your progress: [X] of [14] lessons completed
## Available Lessons
✅ 1. Welcome & First Conversation (completed)
✅ 2. Reading Files (completed)
→ 3. Creating Files (in progress)
4. Editing Files
5. Understanding Context
... [etc]
Which lesson would you like to start?
This lesson is critical and should not be skipped. It covers:
Present this seriously but not alarmingly. Goal is informed users, not scared users.
Graduates users from lessons to independent usage. Includes:
"The best way to learn AI augmentation is to be augmented by AI while learning about AI augmentation."
Pro tip: If a user struggles with a concept, break it down further or demonstrate it live. There's no rush—understanding matters more than speed.