silverwind

X.509 needs to be replaced with a simple key you get for free during domain registration. Then the blast radius of key compromise is a single domain name and not the whole internet.

> Also GitHub should auto-detect the format on the very first “git push” and handle it then

As far as I'm aware, that's not possible because of how the git protocol is designed.

Can't confirm this "pretty generous", many people with lots of contributions to popular projects got any from OpenAI, not even a response to their request.

I disagree, prefer to see directly which forge a module is hosted on.

What's also annoying is that Claude Code now spams "agents-md: no CLAUDE.md found; AGENTS.md loaded" on every startup. I could just do that silently, no need to inform me that I'm using the superior file.

Or a lot more with IPv6.

It's quite a lot wider and thicker than a mini, and imho looks quite ugly when folded, I don't see the appeal as a single-handed device.

Yeah, the UI is the real and only reason to use Nitter. It's a very similar situation with old reddit.

Isn't that basically AsyncGenerator?

It's possible to drive them hard enough to really read everything, but you have to go really hard on them with many many cycles.

Those benchmarks don't tell much, they only check if a problem was solved, not how. Also there's surely a lot of benchmaxxing going on in the model training.

There are no EU instances of GitHub, it's all in the US only.

More importantly: Once Anthropic is gone, all knowlege is lost.

Most models are trained to be as "helpful" as possible which may work for a chatbot but not for code.

Changing output style shouldn't affect thinking at all.

It's sad that such basic stuff took this long. If we're lucky we might even see a `Map` function in our lifetimes.

It's like 100 times slower than ruff. They had to massive caching techniques to get it to a remotely acceptable speed.

On problem of such go linters is they can only lint what the compiler sees so if you have multiple `build` tags, you need to run multiple lint passes. No other language's linter operates like they, they all operate on file level.

Yes, tsconfig is a mess to scope. It needs a re-design to allow glob-based matching for all options and in typescript instead of jsonc.

Seems they not running these agents with the same permissions of the user prompting them, what a disaster.

I think tsdown is much better suited for CLI and library builds. Vite has many web-isms that don't matter for these.

Yeah, a missed opportunity. But I'm sure such a compiler will eventually arise.

True, sometimes a Claude answer is so on the spot, it'd a waste to not post it alongside your short summary.

PR spam is a major problems for repo that run bounties. Maybe GitHub should temporarily block accounts from raising PRs if like 95%+ of them are getting rejected.

Knowing them, I expect a release probably next week, have fun in production with it.

It'll always be a cat-and-mouse game. If npm adds protections, it'll only yield false-positives and workarounds will be trivial.

Almost all these recent compromises seem to involve either cache poisoning or prompt injection via untrusted variables.

Python had these too, no ecosystem is safe.

I think it's just a case of brain drain, followed by reckless AI adoption which both drove the quality down.

All those forks turned out to be inferior projects with substantially less contributions than the originals.