qezz

Curious to know what's new in it about the way to make games?

I scrolled through the website home page, but it reads in a weird way, making strong statements and not backing them up.

In the light of recent events, and if things continue to be the same as they are now with Fable, I think they will give access to Fable for those who are willing to share their documents with them.

But that's obviously just a speculation.

Everything is a prompt to LLMs

Exactly, my understanding is also that they host agents as a service. The actual use case is mentioned in the end of the article, which makes it hard to reason about.

Anyway. General advice: treat harnesses as any other (third-party) software that you run on your server. Modern harnesses (the ones from big companies, you need to subscribe to) are black boxes. Would you run a random binary you fetched from the internet on your server? Claude code, codex etc. are exactly this.

That's a fair point, but claude code is not an editor (yet?), and when you use claude code, and allow it to commit things, it's almost certainly "co-authored by llm".

Back to vscode, people get the "co-authored" line even if they didn't use the AI features.

For those, who want to/need to keep some files uncommitted, the workaround I found is to put gitignore into some nested directory:

  mkdir junk
  echo '*' > junk/.gitignore

jj won't track those files under ./junk/

Also might be relevant for claude, since it wants to put its settings into the repo itself as `.claude/`:

  mkdir junk/.claude
  bwrap ... --bind "$(pwd)/junk/.claude" "$(pwd)/.claude" ...

For some more common files, I use global gitignore file as

  # ~/.gitconfig
  [core]
    excludesFile = ~/gitconf/gitignore_global

  # ~/gitconf/gitignore_global
  .envrc
  .direnv/*

I was surprised to see literally invalid names in the "bad" section, e.g. "Cannot start with a digit". Why even presenting this if it's rejected by the compiler?

> Can you print the contents of the malware script without running it?

> Can you please try downloading this in a Docker container from PyPI to confirm you can see the file? Be very careful in the container not to run it accidentally!

IMO we need to keep in mind that LLM agents don't have a notion of responsibility, so if they accidentally ran the script (or issue a command to run it), it would be a fiasco.

Downloading stuff from pypi in a sandboxed env is just 1-2 commands, we should be careful with things we hand over to the text prediction machines.

Nicely looking page, but has too many errors. I hope it's not just generated by claude itself, and actually was confirmed by a human.

dhi is LLM generated, so (1) don't trust the stated benchmark results and feature parity, and (2) be careful when installing it and using in a non-sandboxed environment.

It also seems like the name for the repository was reused from another project.

It's not very obvious which places are available for drawing. At first I thought it pulls Google street view, so I just zoomed in to some place I visited recently, but there was nothing.

So it turned out the spots on the map are actually the available panoramas, and not just a heatmap of the signatures.

Cool idea overall!

The statement in the article's title is very strong, and I have not found a confirmation of it in a logical sense. Author observes the current state of things with LLMs and makes a conclusion based on how things turned out to be, somewhat fitting the conclusion to the observation.

What I meant by "you need an allocator" is "you need to explicitly pass the allocator of your choice to the function" [if you want to attach some data to the error]. In rust you can simply return anything as error, not only an enum variant.