シャドーイング練習: Learn anything with the /teach skill - YouTubeで英語スピーキングを学ぶ
C1
I realized the other day I've been teaching stuff for 10 years.
181 文
文が短すぎたり長すぎる場合は、Editをタップして調整してください。
1
I realized the other day I've been teaching stuff for 10 years.
2
I was a voice coach for six years and I've been doing this job, teaching devs, for four years.
3
And for a while I've been thinking, wouldn't it be great if I could take everything I know about teaching
4
and put it inside a skill so that anyone could learn anything?
5
I had a long bus ride to London the other day
6
and I wrote a teach skill and it turns out that it's pretty good.
7
It taught me how to solve a Rubik's Cube, which is something I've always wanted to learn how to do, but I never really had the time or inclination to do it.
8
But with this teach skill, it felt like I had a real teacher who was teaching how I like to be taught
9
and was totally aligned with my mission.
10
And I'm going to show you how I did it.
11
The key concept when we're looking at a teach skill is some skills can be stateful and some skills can be stateless.
12
If a skill is stateless, it means that it doesn't retain any state from previous runs.
13
It doesn't have any memory of the things that you've done before.
14
In other words, a stateless skill doesn't save anything on the file system to help it pick up where it left off later.
15
Whereas stateful skills do do that.
16
They save things either to the local file system or they save things to MCP servers.
17
They keep notes that they later track.
18
Initially, I was thinking about teach as a kind of stateless skill where you would just say, teach me this thing and then it would just find some good resources for you
19
and give you like an output that would teach you a lesson about that thing.
20
But I realized that all the good teaching that I do is stateful, where I teach you, I being the teacher, I remember where you've got to, I know what you've learned,
21
I know what you can go to next, and I've got a bunch of great resources
22
that I remember from previous times that I've taught this thing to teach you about it.
23
So I decided that teach had to be a stateful skill.
24
As an example, if you've used my skills repo before, you know that grill me is totally stateless.
25
It just grills you about a topic until you're ready to implement.
26
but then grill with docs is actually stateful.
27
In other words, it saves some local ADRs, architectural decision records.
28
It saves some other stuff, some glossaries to the repo as well.
29
And so grill with docs is stateful and it gets better over time, whereas grill me is totally stateless.
30
Now, one of these is not better than the other.
31
They're just useful in different situations.
32
And so when you're designing skills, you need to be careful to think about whether they need to be stateless or stateful and teach needed to be stateful.
33
The way you install this skill is you go to my skills repo, matpokot skills.
34
You go down to this quick start here and you just run the skills.sh installer and you choose the teach skill.
35
Once you do that, you go to an empty directory and you run teach inside your coding agent.
36
So for instance, in this directory, which is my Rubik's Cube directory, I said teach me how to solve a Rubik's Cube.
37
And you'll notice during my journey how many files it has created here.
38
But the first one it created was the mission.
39
I believe that for a teacher to be effective, you need to understand why a student wants to learn the thing.
40
And so I got this skill to create a mission for you.
41
Here it says, Matt wants to be able to take a scramble three by three Rubik's cube
42
and solve it unaided at least once.
43
The goal is the achievement itself, not speed, not theory.
44
The next thing it does is it creates a set of resources.
45
So it goes and searches the web for actual primary source, high trust resources that it's going to use to create the lessons out of.
46
It does this on the first pass
47
when you ask it to teach you something and then it will continue to update these as you go.
48
And then it will create your first lesson for you.
49
And lessons are stored in the lessons folder and they're numbered like this and they're all individual HTML files.
50
HTML is just so much richer than Markdown.
51
It allows it to be so much more expressive, so much more interactive.
52
And this is really the core of what makes this such a good learning experience.
53
And here is my first lesson, anatomy notation and the white cross.
54
And it basically teaches you just what you need to know at that moment.
55
It gives you diagrams, it gives you very simple explainers, it gives you call outs, and it gives you quizzes too.
56
One thing I find really important whenever you're teaching anything is to develop a feedback loop.
57
And quizzes are okay at this.
58
They're basically good if you can't find any richer feedback loop.
59
Again, if we zoom down this, we've got notations, we've got traps, we've You've got your first skill here.
60
And so it's giving me the knowledge that I need, but also encouraging me to develop the skills I need, which is slightly different.
61
And the skill itself knows that.
62
Once I completed the lesson and I said, yes, I can make the white cross, then it recommended that or rather recorded that inside a learning record here.
63
And these learning records are very simple records that the agent creates when I report how I'm doing.
64
So this allows it to keep track of how I'm doing, just like a real teacher, then tailor the next lesson to what I need.
65
Here's another explainer from slightly further down this course.
66
I suppose it's kind of creating a course as we go based on what I need.
67
So this one here, it was, okay, it started to add like little citations now, again, more quizzes.
68
I mean, I just find this layout really nice to look at and really useful.
69
I was sort of working on the skill as I was building this too.
70
And so I added this little bottom or thing at the bottom.
71
It says it tries to find communities where you can ask questions here.
72
Because sure, you can develop knowledge and skills here.
73
But the only way you're going to develop wisdom about the thing is actually interacting with a community, testing your ideas out in the real world.
74
It also develops reference material for you too.
75
This was something I noticed that I needed.
76
So there's a glossary.
77
This glossary has the anatomy.
78
It has the notation here.
79
It has the solving grip, the daisy sort of anything like weird jargon
80
that we've learned is going to go into the glossary.
81
This is really useful, especially if you're learning like a coding language using this.
82
It also means that future lessons can be more concise
83
because they're able to reference the term in the glossary and you can look back if you're confused.
84
It also creates cheat sheets for you.
85
So this is a solve card here where it's basically just giving me the entire solve.
86
If I want just a single place to look at how to create or solve a Rubik's Cube, this is the place I can go.
87
And finally, just to explain this notes.md file, this is where the agent can note down any of my preferences, note down any watchouts, and it's just kind of internal note-taking for the agent itself.
88
So let's imagine that I sit down to a session
89
and I basically want to fill in the teacher on what I've been doing since I last had a session.
90
So I just say teach and I can basically complete the Rubik's Cube now except except for the corner cycle, which I've still not quite got into memory.
91
Let's go ahead and see what it says.
92
It starts, of course, because this is an empty context by checking the state of my teaching workspace.
93
It's checking all of the solve cards, seeing how far I've got to.
94
It's now given me a decent diagnosis for what I'm feeling.
95
It's saying the concept is solid for the corner cycle, which is a particular algorithm, but it's the one that hasn't reached muscle memory.
96
Let me look at the existing memorization lesson and one earlier lesson to match the house style.
97
So we've turned the agent basically into a teacher
98
and you've got to work with it in the same way that you would any one-to-one teacher.
99
You tell it what problems you're having and it will design lessons to find solutions.
100
The model I'm using here is Opus 4.8 with medium effort, by the way.
101
It's also using a key teaching term here, the zone of proximal development.
102
One algorithm, pure muscle memory work.
103
The zone of proximal development is this key teaching idea which I am obsessed with.
104
And it's the idea that you should always teach someone in the area where they are perfectly challenged but not intimidated.
105
This means that every lesson has to be concise, compact and exactly framed at that zone of proximal development so the student's not bored but also not freaked out.
106
So we can see it's now created this lesson, the corner cycle lesson.
107
And it's created this exact lesson for us.
108
So it's basically just breaking it down, paying more attention to it, giving me new mental models.
109
One four-move phrase played twice.
110
Okay, that is handy.
111
And The idea is I would walk through this lesson.
112
Oh, look at this.
113
Oh my gosh, look at this.
114
So it's actually given me an interactive tap thing I could do.
115
Okay, so it's U R U prime L prime U R prime U prime L.
116
Oh my God, look at this.
117
So it's got guided mode.
118
So I can turn off that.
119
Oh, that's unreasonably cool.
120
You see what being in HTML gives us here, right?
121
We've got the full power of the browser to mess about with here.
122
So you get the idea of what these lessons look like.
123
They are short, they are really focused, and it gives me exactly what to practice now.
124
And this also means because everything has been saved to the file system, I can clear this, and every time I run teach again, it will have all the context it needs to keep me in the zone of proximal development.
125
Let's briefly look at the skill itself here.
126
It's really quite simple, although it is one of the longer skills that I've put together.
127
The user has asked you to teach them something.
128
This is a stateful request.
129
They intend to learn the topic over multiple sessions.
130
It goes into the shape of the teaching workspace, which we've looked at already, and it has a section on philosophy here.
131
So I really define the terms that the AI should think in.
132
It should think in terms of giving them knowledge from high quality, high trust resources, skills, highly relevant interactive lessons and actually developing the skills.
133
And then wisdom that comes from interacting from other learners and practitioners.
134
We go further down here and we touch on the shape of the lessons themselves.
135
We've talked about the mission as well, mission.md, the zone of proximal development, acquiring knowledge and skills, how to do that,
136
acquiring wisdom, which basically when the learner is ready to go out into the world, once they've got the knowledge and skills, how do you delegate them to the community?
137
When the user asks a question that appears to require wisdom, your default posture should be to attempt to answer, but to ultimately delegate to a community.
138
My idea with this teach skill is not to have you hooked onto the agent for learning everything.
139
It's to actually give you the skills you need to feel confident to go out
140
and work with a community and send you out into the world.
141
That's the dream of this stuff.
142
We've also got more material here on reference documents, then finally on notes.md.
143
But that really is the skill.
144
There's a few little bits of supporting documents for like learning record format, the mission format, the resources format, etc. In terms of engineering,
145
I can imagine this skill being really useful in onboarding people to a code base.
146
Creating documentation is a real pain in the arse because not only do you need to keep it up to date, but that documentation is probably outside the person's zone of proximal development.
147
They might have worked with your stack before.
148
They might just need to know the specific problem domain or they might have worked with the problem domain before, but they have no idea what TypeScript is or things like that.
149
So you can think of teach here as, you know, you start them off in their own workspace.
150
You point them at the code base.
151
You teach them how to, you know, they learn independently how to work with the code.
152
what the concepts are, and guess what?
153
You've got a productive employee in record time.
154
For me, I'm excited to use this just on my side projects, on having fun, on just learning stuff that I wouldn't have learned how to do before, you know, solving this freaking Rubik's Cube.
155
I might even dig out a bit of old chess material and see if it can teach me some openings.
156
And I wanted to end here by talking about an idea
157
that I've had for a while that I think this teach skill really proves or rather makes very plain to me, which is that we, the developer community,
158
engineers out there, are the first people to really experience what AI can do on something that it's really good at.
159
AI is currently better at writing code than it is at almost anything else.
160
There are a few exceptions, but really AI is very good at writing code
161
and developers have an advantage
162
that we're the first people to really get to test AI out on a problem where it's very good at it.
163
This means that we are the first people really in the world, with a couple of exceptions, that get to know AI, that get to build these cool things.
164
And it means that we're the first movers in this new space.
165
We can take the ideas that we develop here, turn them into skills and start bringing them out to the world.
166
I think that's incredibly exciting.
167
And it's something that I'm going to be exploring more.
168
And this teach skill is kind of the first phase of that.
169
So when you're working with Claw Code or Codex and you're struggling with it and you're losing motivation, you think, God, this sucks.
170
Just think that you'll be able to take the skills that you are learning now, the instincts that you're building, working with AI and take them outside of coding domains.
171
That, I think, is going to be an incredibly valuable skill, no matter what the future of work looks like.
172
We're all going to be working with AI and we are the first people to really get to do it.
173
Now, if you're digging my skills, then you should go to AI hero forward slash skills
174
and sign up for my newsletter that lets you know whenever I release a new skill, whenever I have any changes, because I do change the skills all the time.
175
If you want up to date information on how to use my skills for engineering and for other stuff, too, then check it out.
176
But overall, thanks for watching.
177
I can't wait to see the languages that you're learning, coding languages or human languages, the new skills that you're developing.
178
I'm going to get this thing to teach me vocal harmonies as well.
179
It's always a skill that I've wanted to know but never been very good at.
180
And I just can't wait to get using this thing myself and seeing how you're using it too.
181
But overall, thanks so much for watching and I will see you very soon.
アプリをダウンロード
話したすべての文をAIが採点
TRENDING
人気動画
この動画でスピーキングを練習する理由は？
この動画は、教育スキルを活用して新しいことを学ぶ方法について話しています。特に、話す力を向上させたい英語学習者にとって、この内容は非常に役立ちます。スピーキングの技術は、言語を習得する際に欠かせない要素であり、動画内での具体的な学びの過程を追うことで、実際の会話における適切な表現や流暢さを模倣することが可能です。こうしたスキルは、shadowing siteを利用する際にも非常に効果的です。実際にネイティブスピーカーの発音や抑揚を真似ることで、IELTS スピーキング対策にとっても非常に有益です。
文法と表現の文脈
以下は、動画で使われた重要な文法構造と表現のいくつかです：
- will + 動詞: 将来の意図を表現する際に使用されています。「I will teach you how to solve a Rubik's Cube」という表現は、具体的な行動計画を示します。
- be + 動詞のing形: 継続的な行動や状態を表すために使われています。「I’m writing a teach skill」では、現在進行形を使って、現在のアクションを強調しています。
- 過去形: 経験を語る際に使用されています。「I realized that I've been teaching stuff for 10 years」というフレーズは、過去の経験に基づいた現在の理解を示します。
- if節: 条件を提示するために進行形が使われています。「If a skill is stateless」という部分では、特定の条件に基づく違いを説明しています。
一般的な発音の罠
この動画では、いくつかの発音の難しい単語やフレーズが使用されています。特に注意すべき点は以下の通りです：
- skill: 発音が難しいため、正確に言えるようになることが大切です。
- Rubik's Cube: 名称の特異性ゆえに、発音が混乱することがあります。練習する際は、正確な音を確認しながら練習しましょう。
- resources: この単語は複数形であるため、発音に注意が必要です。
これらの要素を意識しながら、shadowspeaksやshadow speechを活用して練習することで、より効果的にスピーキングのスキルを向上させることができます。実際の会話に近い状況を想定し、自身のスピーキングを反復し、改善していくことが重要です。
シャドーイングとは？英語上達に効果的な理由
シャドーイング（Shadowing）は、もともとプロの通訳者養成プログラムで開発された言語学習法で、多言語習得者として知られるDr. Alexander Arguelles によって広く普及されました。方法はシンプルですが非常に効果的：ネイティブスピーカーの英語を聞きながら、1〜2秒の遅延で声に出してすぐに繰り返す——まるで「影（shadow）」のように話者を追いかけます。文法ドリルや受動的なリスニングと異なり、シャドーイングは脳と口の筋肉が同時にリアルタイムで英語を処理・再現することを強制します。研究により、発音精度、抑揚、リズム、連音、リスニング力、そして会話の流暢さが大幅に向上することが確認されています。IELTSスピーキング対策や自然な英語コミュニケーションを目指す方に特におすすめです。