- > As of Dec 2025, Sonnet/Opus and GPTCodex are both trained and most good agent tools (ie. opencode, claude-code, codex) have prompts to fire off subagents during an exploration (use the word explore) and you should be able to Research without needing the extra steps of writing plans and resetting context. I'd save that expense unless you need some huge multi-step verifiable plan implemented.
Does the UI shows clearly what portion was done by a subagent?
- Oh ok, thanks.
It would be so cool if LLMs could get the type of a variable when it's unclear (specially in languages with overloading and whatnot). Or could get autocomplete if they get stuck with a code. Really I think that agents and LSP should be hybrid, and maybe the agent could inform the LSP server of some things like things to warn (IDE diagnostics could be driven by a combination of LSP and AI agents)
- Hi, I just looked up and two weeks ago someone made this suggestion in Cursor forum
https://forum.cursor.com/t/support-of-lsp-language-server-pr...
> Feature request for product/service
>
> Cursor IDE
>
> Describe the request
>
> It would be a huge step up if agent could interact with LSP (Language Server Protocol).
>
> It would offer :
>
> renaming all instances of a symbol over all files in one action
> quick navigation through code : fast find of all references to a property or method
> organize imports, format code, etc…
And last Friday a Cursor engineer replied "Thanks for the idea!"
So how does the AI agent in Cursor currently have access to LSP?
(I am most interested in having the agent use LSP for type checking, documentation of a method call, etc. rather than running slower commands)
(note, there is an open PR for Zed to pull LSP diagnostics into an AI agent thread https://github.com/zed-industries/zed/pull/42270 but it would be better if agents could make arbitrary LSP queries or something like that)
- How does an electric shock throws someone across the room? What's the mechanism for this push?
I know a shock can paralyze (by contracting the muscles) and it can burn (by joule effect) but never seen one push
- The problem this (and Dioxus Native) solves is that someone might prefer Rust anyway.
Most of times it's just a personal preference, but sometimes it's due to using Rust libraries or already having code written in Rust that can be reused. There is Rust <-> Kotlin FFI (also Rust <-> Dart) but sometimes people don't like it
- What IDE agent gets access to LSP?
I use Zed and unless there is some MCP server that provides the same thing as the LSP server, the Zed agent won't have access, even though it's in an IDE that supposedly has this information
- Sometimes, monetization is just impossible. But if you insist, have the skill call an API that needs a token, and charge for the API
- curiously other pages with three.js demos at the landjng page like svelteflow.dev/ doesn't have this issue
i think the trick is to not attempt to render at each frame (that is, don't try to provide something similar to a game engine). just render it based on user input, like the page I linked
- I thought that no other browser engine could be provided on iOS. so no ladybird's engine, no servo, no gecko, no blink, only webkit
- > myrocks
anything like this, but for postgres?
actually, is it even possible to write a new db engine for postgres? like mysql has innodb, myisam, etc
- Even with overcommit enabled, malloc may fail if there is no contiguous address space available. Not a problem in 64 bits but may occasionally happen in 32 bits
- Whoa, that's cool!
What do you think about Typescript? I mean it's unsound but - it sounds like an ok compromise
- Your assumption is that a 11 second delta is a somewhat better evidence than a 90 seconds delta, but the provided article successfully defended this isn't the case IMO. It depends on the last activity of the user
The article also shows that there's a 40 second delta between the harassing account and the harassed person himself, further semonstrating this doesn't mean anything and can happen purely by chance
- The alternative is no reconstruction at all on one hand, and adding fake detail on another hand
And if one wants to add fake detail, why should archeologists be involved? Just have AI generate them
- You can code without it. Maybe you don't want to, but if you're a programmer, you can
(here I am remembering a time I had no computer and would program data structures in OCaml with pen and paper, then would go to university the next day to try it. Often times it worked the first try)
- I mean, is there some open source project that already uses webhooks to replace Github Actions?
Rather than having to write some ad hoc code to do this
- Will migrating to annotate-snippets change rustc/cargo formatting of errors in any way?
Also, in what sense it is more conservative?
- There was a system with dependent types that ruled out invalid html at compile time, even dynamically generated html (rather than a runtime error, you would get a compile error if your code did something wrong)
https://github.com/urweb/urweb
http://www.impredicative.com/ur/
Needless to say it wasn't very practical. But there was one commercial site written in it https://github.com/bazqux/bazqux-urweb (the site still exists but not sure if it's still written in ur/web)
Also, what an horrifying way to die