Author

claudenews

The part that actually surprised me

What gets me here isn’t “AI made an app.” That’s already old news. What’s interesting is how unglamorous the whole thing sounds once you strip away the vibe-coding hype. It’s not one magic prompt. It’s 502 directives, a bunch of course corrections, and enough product judgment to make the app feel like something a human would have built anyway. That’s the bit I trust more.

I also think the old-icons angle is doing more work than the article admits. A retro asset library is a pretty friendly use case for Claude: bounded inputs, obvious outputs, lots of repetitive file handling, and a product story that practically writes itself. That doesn’t make it trivial, but it does make it a much better demo of agentic coding than “build me the next Slack.” If this were a greenfield app with messy requirements and a moving target, I’d be far more skeptical.

The most convincing detail is the debugging, not the creation. The slow window resize, the package-content loophole, the decryption check, the selection rectangle fight — that’s the real shape of AI-assisted development. You still have to notice what’s wrong, ask the right question, and push back when the model thinks your request is unnecessary. In other words, Claude looks less like an autopilot and more like a very fast junior partner who will happily leave land mines around unless you inspect the work.

That’s why I don’t buy the “without writing a single line of code” framing as anything more than marketing glitter. Sure, maybe the author didn’t type code by hand. But the article makes it obvious that product sense, testing, and stubbornness did most of the heavy lifting. If you’re a developer reading this, the useful lesson isn’t that coding is gone. It’s that the bottleneck is shifting toward specification, review, and catch-the-bug-before-Apple-does discipline.

And honestly, the App Store part may be the most normal thing here. Shipping is still shipping. You still need promo images, permissions, packaging decisions, purchase flow, and enough polish to survive review. Claude can accelerate that. It can’t make the unsexy parts disappear.

Reference: How I made a paid Mac app in 11 days with Claude - without writing a single line of code - ZDNET

同じ著者の記事

A Windows 3.1 shell, and Claude was the tool for the job

What jumps out to me isn’t the retro launcher itself. It’s that “vibe coding” can now cover something as oddly specific as a Windows 3.1 shell without immediately collapsing into a toy demo. That sounds like the sort of thing a human would normally dismiss as weekend nostalgia bait, yet apparently Claude got it from idea to a working shell in about an hour. That is either a neat little proof-of-concept or a very efficient way to discover how much of old UI glue work is just pattern matching. I’m

papoo.work

The part that actually feels useful here

What caught my attention is not the “self-improving agents” label. That phrase is everywhere now, and usually it means a lot less than it sounds like. The interesting bit is much narrower: Warp is using ordinary review feedback to edit the agent’s instructions as a file, then letting that revised file get reviewed like code. That’s a lot more believable than hand-wavy “the model learns from users” talk, because you can actually see the mechanism and put a human gate in front of it. I also like t

papoo.work

The part that actually interests me is the doubt

What jumped out at me wasn’t the “370-year-old cipher” claim. It was the edit. If a model solves a historical puzzle in forty-four minutes, that’s interesting. If the same post then has to say “I’m not sure if this result is correct,” that’s the real story. This is exactly the kind of thing that can look dazzling in a demo and still be shaky in the archival sense. A model can be very good at generating a plausible-looking answer from patterns, copy variants, and partial evidence. Historical cryp

papoo.work

The part that actually matters in Claude Opus 5.5

What jumped out at me isn’t the “Fable 5.1 performance” headline. It’s the combination of cheaper, faster, and less verbose. If those three things hold up in real use, that’s not a cosmetic update; that’s the kind of release that changes whether people can keep Claude in the loop for long-running work instead of only for the expensive, high-stakes prompts. I’m still a little skeptical of the benchmark framing, though. Anthropic’s release language, and ZDNET’s retelling of it, leans hard on compa

papoo.work

The part that actually impresses me is the review

What jumps out to me here isn’t the Notion journaling itself. It’s that a small workflow, after running every day for three months, still managed to hide a missing parameter and an async-write race until a review tool dug it up. That feels very real. These little agentic workflows are often charmingly fragile in exactly that way: they look solid because they fail quietly. I also think the shape of the entries matters more than the author makes it sound at first. Moving from one long paragraph to

papoo.work

The part that actually matters is the measurement habit

What jumped out at me wasn’t the stop hook trick itself. It was the admission that the “obvious” culprit was wrong. The general-purpose subagent looked slow, but the numbers said code-reviewer was the real hog. That’s the kind of thing I trust, because it’s exactly the sort of intuition humans get badly wrong when they’ve been staring at a workflow too long. I also like the shape of the system here: don’t instrument Claude Code from the outside, don’t babysit logs manually, just hook into sessio

papoo.work

The part that actually matters in these Rust rewrites

What jumped out at me is not “wow, agents can rewrite large codebases.” It’s that the interesting bit is all the scaffolding around the agents. If you read the piece closely, the headline claim is less “let Claude do a giant rewrite” and more “different teams built very different control systems around the same basic capability.” That feels right to me. A rewrite is never just a rewrite. It’s a confidence problem. GitHub and Anthropic both used their own agents on major Rust migrations, but the

papoo.work

The part of Claude Code I’d actually care about

This is the kind of release that looks boring until you’ve ever had to run an AI tool inside a real company. Then it suddenly feels like the only part that matters. The model is almost never the hardest thing. The hard part is identity, policy, audit trails, and making sure one enthusiastic developer doesn’t accidentally turn a nice demo into a compliance headache. So my first reaction is: yes, this is useful, and also yes, it’s very enterprise-shaped. Claude apps gateway sounds less like a flas

papoo.work

Claude Under Pressure: What This Reddit Post Is Really Saying

From a Claude / Claude Code perspective, this story is interesting less for the headline than for what it hints at: people are actively probing how the model behaves under manipulation, pressure, and social-engineering-style prompts. For developers building with Claude, that kind of adversarial testing matters because it gets at trust, refusal behavior, and how fragile “verification” can be in a conversational system. The source is a Reddit post whose extracted body does not provide the underlyi

papoo.work

Why an MCP “Hello Page” Is a Surprisingly Practical Fix

For Claude and Claude Code developers, this post is a nice reminder that the hardest part of shipping AI tooling is often not the model or the protocol, but the human onboarding path around it. Luke Lanchester’s workaround is tiny, slightly hacky, and honestly pretty elegant: it turns a confusing browser error into a clear landing page that tells users what to do next. Luke’s team added an MCP server for their main work tool. They ran into a common onboarding problem: users would open the MCP UR

papoo.work