ฝึกพูดภาษาอังกฤษด้วยเทคนิค Shadowing จากวิดีโอ: Can Cursor's HARDCORE Review Skill Stop The Slop?

Automated code review is one of the most impactful ways that you can improve the code quality coming out of your agent.

I've known this for a while, but it's taking me a while to kind of implement it

and figure out a reusable skill that I can give to people to review their code.

I've got this review skill here in my skills repo, which is currently sitting at 109,000 stars,

and it is currently marked as in progress.

I'm sort of okay with it, but I'm not terribly happy with it.

So I've been looking around for inspiration in other skills that I can copy from, steal ideas from, and one crossed my path that I want to show you.

It is this one from the Cursed team.

It is the Thermonuclear Code Quality Review.

Use this skill for an unusually strict review focused on implementation quality, maintainability, abstraction quality and codebase health.

And one thing I think is notable about this skill is how ambitious it asks the reviewer to be.

asking it to be very ambitious and look for code judo moves throughout the review.

The skill itself is simply one file.

It's just a skill.md up here.

And what I thought I'd do is I would copy it to my local system, try it out on some actual code of mine, and see what it comes up with.

Yesterday, I spent a lot of time working on Sandcastle, my open source software factory.

And so I figure I would review the last X number of commits and see what it thought about them.

So I'm going to be pretty loose here.

I'm just going to say thermonuclear code quality review.

Review the last five PRs that made it to main.

I'm going to stick it on auto mode and

while it's doing this let's go and actually read the skill because that should explain the skill a bit more.

So it starts from this baseline.

Perform a deep code quality audit of the current branch's changes.

Rethink how to structure implement the changes to meaningfully improve code quality without impacting behavior.

Work to improve abstractions, modularity, reduce spaghetti code.

Improve succinctness and legibility. Be ambitious.

If there is a clear path to improving the implementation that involves restructuring some of the code base, go for it.

Be extremely thorough and rigorous.

Measure twice, cut once.

What I've often found with review skills like this is that the agent is not ambitious enough.

If you pass an agent a diff, then it will usually treat that diff as its bounds within which it can work.

Whereas this prompt appears to be going beyond that.

It's essentially saying look throughout the entire code base for opportunities, but starting from this current branch's changes.

It also goes on to add a bunch of non-negotiable additional standards.

Be ambitious about structural simplification.

Again, the ambition.

Do not let a PR push a file from under 1K lines to over 1K lines without a very strong reason.

This is really interesting.

I've actually reached this conclusion myself as well large files are just quite hard for agents to navigate

because they need to ingest the entire file into their context

window in order to find the thing that's actually useful within it.

A much better way to structure that is to split them into multiple files

and let the file name of the file be the context pointer

that tells it what's in that file and whether it might need to open it.

This ends up being a lot more context efficient.

I have generally split my files if if they go over 5K tokens, but this 1K lines is sort of, I guess, a similar rubric.

Do not allow random spaghetti growth in existing code.

Okay, I see, It's sort of arguing against nesting here.

If a change adds weird if statements in random places, treat that as a design problem, not a stylistic knit.

Prefer pushing the logic into a dedicated abstraction helper state machine policy object or separate module instead of tangling an existing path.

Interesting.

This is another way of telling it to be aggressive about, you know, if there's a bunch of nested if statements and weird conditionals,

maybe abstract that into a cleaner abstraction or a helper or something.

It's arguable whether I prefer that.

I suppose sometimes I do, sometimes I don't.

But let's assume it's a good thing for now.

Bias towards cleaning the design, not just accepting working code.

Again, pushing it to be ambitious.

Prefer direct, boring, maintainable code of a hacky and magical code.

This is like a classic one in these prompts.

This comes from, I think, Simplify in Claude Code.

Not the same wording, I think, but a similar idea that you want simple, direct code that's easy to read.

I really like this one, actually.

Push hard on type and boundary cleanliness when they affect maintainability.

So we're specifically talking about types here.

Question unnecessary optionality, unknown, any, or cast heavy code when a clearer type boundary could exist.

This is kind of TypeScript focused here.

Unknown and any are specifically TypeScript terms.

And the unnecessary optionality is one that always gets me.

Whenever an agent adds a prop onto a React component, let's say, it always adds it as optional.

I don't know why.

It's so stupid.

Even when it's always required, it will add it as optional just to make it backwards compatible or something, or to lessen the blast radius of the change.

So yeah, question unnecessary optionality is a great one.

Keep logic in the canonical layer and reuse existing helpers.

Prefer existing canonical utilities and helpers over bespoke one-offs.

Yes, I suppose it's basically just telling it to look for

places where this has already been solved in the code base and use those instead.

Makes sense.

Treat unnecessary sequential orchestration and non-atomic updates as design smells when the cleaner structure is obvious.

If independent work is serialized for no good reason, ask whether the flow should run in parallel instead.

I see.

This is about performance, essentially.

Obviously, when two things that are independent, if they run in parallel, then it's going to be faster than if they run sequentially.

So that's kind of what it's going for here.

But it's also saying do not over-index on micro-optimizations.

Okay, so it's basically telling it don't go too far.

I think if this was my skill, I would definitely rewrite this to be a lot more direct.

Treat unnecessary sequential orchestration, non-atomic updates as design smells.

That's just word salad to me.

I don't know what that means.

So this is really cool.

Primer review questions.

For every meaningful change, ask, is there a code judo move that would make this dramatically simpler?

That's great.

I love that.

Can this be reframed so that fewer concepts, branch or helper layers are needed?

100

Lovely.

101

I don't love this.

102

Does this improve or worsen the local architecture?

103

You've got to say exactly what good

104

and bad looks like to an agent in order for improve or worsen to mean anything.

105

Overall, this set of questions, along with the kind of rules above, give the agent a nice kind of way in to talking about the code, which is what you need.

106

And now it talks about really bad stuff.

107

Escalate findings when you see a complicated implementation where a cleaner reframing could delete whole categories of complexity.

108

Refactors that move code around but fail to reduce the number of concepts a reader must hold in their head.

109

Yeah, there's a bit of repetition going on here unnecessary casts, any unknown or optional params.

110

What sort of scares me about these big review-based prompts is

111

that this is a huge ball of mud for the agent to read.

112

Like, there's a lot of instructions in here, and it's hard to know what to prioritize for the agent.

113

So I don't know.

114

This makes me a little bit nervous.

115

I do like this, though.

116

When you identify a code quality problem, perverse suggestions like delete a whole layer of indirection rather than polishing it.

117

Again, ambition.

118

Split a large file into smaller focused modules.

119

Again, you know, making things easier to navigate for the agent.

120

Again, duplication.

121

Make type boundaries more explicit so the control flow gets simpler.

122

There's a lot of duplication throughout a lot of this.

123

This could be cut down, I think, quite a lot.

124

Review tone.

125

I don't know why this is here.

126

This is just sort of saying choose your tone, I suppose.

127

Be direct, serious, and demanding about quality.

128

Do not be rude.

129

This seems like a crazy thing to add to a skill.

130

I don't know why that's here.

131

What this does do is it does really punch the language that the agent should be using.

132

So we're really emphasizing code judo, saying decompose, pushes the file past, makes the surrounding code more spaghetti.

133

I like that.

134

But the down we can say output expectations.

135

Right.

136

This is nice.

137

It's saying to prioritize findings in this order.

138

It's saying to float the important stuff to the top and legibility and maintainability concerns are at the bottom.

139

Structural code quality regressions right at the top.

140

Right.

141

And it is asking for an approval here.

142

So it's approving or rejecting the PR.

143

And again, tons and tons of repetition here.

144

This skill could be a lot shorter.

145

So what we have is a large block of text that basically says be more ambitious.

146

Here are some specific things that you can focus on in your review.

147

Really go nuts here and propose a ton of structural changes.

148

Make sure that you prioritize your findings in a certain order so you don't flood it with useless crap.

149

And then approve or reject based on these conditions.

150

What I don't like here is there's no mention of testing.

151

There's no mention of scenes.

152

There's no mention of any kind of improving the feedback loops to make future runs better,

153

which in my view is the entire point now of having a good code base

154

or having a code base that's easy to change and modular and easy to navigate.

155

All of this appears to be focused on actual source code, none of it on tests.

156

Interesting.

157

But okay, let's read what it said here.

158

So it's taken the last five PRs to main and it has found some blocker class structural issues.

159

Okay, it's found that an init service is now a big file, so it's over 1,000 lines and it mixes a bunch of stuff here

160

and it should have been preceded by a split and it's proposed a nice split there.

161

Oh, it's also trying to create an abstraction here,

162

a little make registry generic function returning this would delete 20 lines of duplicated boilerplate at the same time.

163

Feels good.

164

That's nice.

165

So we now go to the next one, the feature specific if issue tracker name custom scattered across three layers.

166

Interesting.

167

It's basically saying that instead of this being a special case if statement here, we should instead do a bit of code judo

168

and push the custom tracker variations into a type itself and then it can be read later.

169

I think in terms of suggestions here, I happen to know this code quite well.

170

I think we are at two out of two here.

171

That seems like two really good suggestions.

172

Down here we have an inconsistent contract.

173

Template args carries both shell commands and prose markers.

174

It's basically saying that some of these are runnable but some of these are not runnable here.

175

And it's saying that maybe we should widen the type to either a command

176

or a to-do marker discriminated union or use a different field entirely for unfilled markers.

177

So basically, it's basically trying to strengthen the type boundary here

178

so that we don't later pass in a prose marker into something else.

179

That's interesting.

180

I think that this comes from an inaccurate understanding of the whole system, which is okay.

181

You're going to get some false positives, I suppose, to the false positive in any review prompt.

182

So this is the kind of thing, if it came up in a PR, I would say, this is fine.

183

Don't worry about it.

184

So two out of three, not bad.

185

Let's look at the strong code quality issues.

186

Aha, we do have a weird bug here that it's, well not a bug, just a weird bit of code design.

187

We essentially have different templates in Sandcastle that each declare the dependencies that they need.

188

And mostly they declare Zod as their dependency.

189

So we have this weird code path in here that looks like it just hardcodes Zod.

190

And then, interesting.

191

Yeah, overall this is quite hard to explain, but it's definitely pulled up something weird here.

192

So I think we're at three out of four, which is good.

193

Oh, it's found some swallowed errors here.

194

Exact sync inside effect.sync with swallowed errors.

195

Interesting.

196

We can see it's trying something inside here.

197

And if it fails, then it just like returns false inside here.

198

So yeah, this is definitely another thing that I would like the reviewer to look at.

199

Looks like we started decomposing a large file into small files, but only half finished.

200

So this again is a good one.

201

This is five out of six so far.

202

And it's now saying there is a bit of prompt duplication within some prompts that were changed here.

203

So the change is byte identical for two different prompts here.

204

It's saying that we should refactor those into an issue list preamble.

205

I don't think that's right.

206

I think that prompts should just be independently changeable.

207

So not bad though.

208

Five out of seven.

209

Then it's got a list of smaller items worth fixing here.

210

I've done a quick scan and I would say most of those look pretty good and interesting.

211

I'm kind of intrigued by the approval bar here.

212

Under the skills stated bar, a couple of the PRs should not have landed in their current shape.

213

The behaviour is correct in all three substantive PRs, but the code base is meaningfully messier than it was a week ago.

214

Well, cool.

215

I mean, we got some really good feedback from this skill, I think.

216

I think what this is teaching me is

217

that actually getting the review to be super ambitious and getting it to push a lot of different options

218

will give you more false positives, but those false positives are pretty easy just to say no to, right?

219

It's the ones that you miss that you never know about, the opportunities for improvement that you never see, those are the dangerous ones.

220

Overall, I will clean up this skill so it's not quite so duplicative, so there's a bit more dry.

221

And I would also just get it to focus a lot more on tests as well.

222

Think about the seams in your codebase, kind of like what my improved codebase architecture does.

223

But overall, I think this is worth pulling down, experimenting with, and just seeing what comes out of it.

224

Now, if you dig this stuff, then I'm running a cohort starting next week, starting June 1st, on AI coding for real engineers.

225

This has been my most subscribed to course ever.

226

People are going nuts for this.

227

We're going to have, I think, around 4,000, 4,500 people in there.

228

So yeah, it's absolutely wild.

229

But if you're enjoying this stuff, if there's a skill that you want me to review, I really like making that content because it lets me steal ideas from other people's great skills, then let me know.

230

Nice work, and I'll see you very soon.

เทคนิค Shadowing คืออะไร?

Shadowing เป็นเทคนิคการเรียนรู้ภาษาที่ได้รับการรับรองทางวิทยาศาสตร์ พัฒนาขึ้นสำหรับการฝึกนักแปลมืออาชีพ วิธีการนี้เรียบง่ายแต่ทรงพลัง: คุณฟังเสียงภาษาอังกฤษจากเจ้าของภาษาและพูดตามทันที — เหมือนเงาที่ตามผู้พูดด้วยช่วงเวลาห่าง 1-2 วินาที การวิจัยแสดงว่าเทคนิคนี้ปรับปรุงความแม่นยำในการออกเสียง ทำนองเสียง จังหวะ การเชื่อมเสียง การฟังเข้าใจ และความคล่องแคล่วในการพูดได้อย่างมีนัยสำคัญ

ฝึกพูดภาษาอังกฤษด้วยเทคนิค Shadowing จากวิดีโอ: Can Cursor's HARDCORE Review Skill Stop The Slop?

ทำไมการฝึกพูดกับวิดีโอนี้ถึงสำคัญ?

ไวยากรณ์และวลีในบริบท

กับดักการออกเสียงที่พบบ่อย

เทคนิค Shadowing คืออะไร?