Mentions

I made a data-only catalog of iOS libraries for AI agents (stars, last release, min iOS), refreshed weekly Coding agents pick iOS dependencies from training data, so they can't tell whether a library still ships releases or what iOS version its current release needs. This repo is a catalog of 91 iOS libraries and tools in 55 categories. Each one shows the same fields: stars, latest release, last commit, and min iOS and Swift tools version read from Package.swift at the release tag. A GitHub Action refreshes the data weekly. The top pick in each category comes from a fixed rule: drop archived repos and those with no release in 2 years, then sort by stars. Provider SDKs (backend, payments, analytics) are listed but not ranked. Each category also names the Apple API that does the same job. Examples of libraries with no release in over 2 years: Charts (28k stars, last release Feb 2024), Moya (15k, Aug 2022), KeychainAccess (8.3k, Mar 2021). There's also an agent skill. It asks what the app needs, applies the rule above, creates the project with one of four recipes (XcodeGen, Tuist, MobileBuildMCP, Xcode MCP; each run end to end on Xcode 27.2), then builds and launches it in the Simulator. jeffy-w/awesome-ios-for-agents No opinions in the list: descriptions only say what a library does, and notes quote a source. If you think a library is missing or a category is wrong, issues and PRs are welcome.

Full video walkthrough of my "AI for Software Architects: The Mental Map" diagram! 🎥👇 In 2 minutes, I break down how LLMs, Context Windows, Vector Embeddings, RAG & AI Agents map 1:1 to Distributed Systems patterns: vishwanath-tns/ai-for-software-architects t.co

Karukan IMEのWindows用フロントエンドを追加しました kachaya-ime/karukan rustで実装しています インストーラーまで作ってくれるので楽でした ライブ変換の様子がわかるようにゆっくり目に打鍵しています t.co

DiscImageStudio: burn visible artwork onto CDs, DVDs and Blu-ray discs JiaFeiMiao-K-Cat/DiscImageStudio (news.ycombinator.com)

Asset Maintenance and Monitoring System. A system for managing assets in a fictional manufacturing company. Official website link: amms-f1sg.onrender.com GitHub source code (Read LICENSE.md for usage): khianvictorycalderon/Asset-Maintenance-System t.co

diffbandit.nvim v0.3.0: side-by-side diffs that don't pad your files, now with bezier connectors in Ghostty/Kitty diffbandit.nvim is a diff and merge viewer for Neovim. It works more like IntelliJ's diff view than :diffthis. Both files keep their own shape and line numbers, with no filler lines or stretched rows. A narrow gutter between them draws how hunks connect. Each pane scrolls on its own, and the connectors follow whatever is on screen. AI disclosure: much of the code was written with Claude Code, grok build, or codex under my direction and review. What it does: - File, buffer, Git worktree and branch/commit diffs. The right side is a normal editable buffer, so LSP, completion and :w work as usual. - Hunk stage, unstage and discard, plus a commit panel with a live diff preview. - A 3-way merge resolver (local/result/remote) with >>/<</gb accepts, gA to apply every non-conflicting change, and resolve-on-write staging. - Folder diffs, overview gutters, and a hex view for binary files. New in v0.3.0 - Bezier ribbon connectors in Ghostty, Kitty, Rex, and anything else built on libghostty. They use Kitty graphics with Unicode placeholders, so scrolling never rewrites buffer text. The renderer is picked automatically on Neovim 0.12+. tmux and other terminals keep the Unicode cell renderer. - Standalone launcher. diffbandit (or git diffbandit) opens every changed file in the repo, lazygit-style, and takes the same arguments as git diff. - Merge fixes. gA used to place changes at the wrong rows after a partial resolution, which could overwrite lines. It's fixed, as is tracking of conflict regions after partial accepts. Repo: CoreyKaylor/diffbandit.nvim Release notes: github.com/CoreyKaylor/diffbandit.nvim/releases/t… Feedback welcome, especially from people on Kitty or Rex, since I've mostly tested in Ghostty.

LunarArc Mobs of Mayhem (1.21.11) is out now! Download the initial release: github.com/LunarArcDevs/LunarArc/releases/tag/Mob…

我运营着好几个站，最后还是自己写了个 CMS 我运营着好几个站，最烦的不是写稿，是每个站各装一套后台、插件和更新。最后还是自己写了个 CMS：一个后台管所有站，发布即静态，AI 写作在编辑器里，一条命令自托管。 一个后台管所有站，不是一套系统管一个站。运营几个内容站的人最痛的不是写稿，是运维：5 个站就是 5 套后台、5 套插件、5 份更新和备份，改一个全站通用的东西要登 5 次后台点 5 遍。AI-CMS 把站点、栏目、内容、媒体、版式收进同一个后台，带 RBAC 权限和回收站，加新站是点几下的事，不是再装一套系统。版式那层也做到了可管：HTML 模板用 HY_ 标签，带版本管理和资产依赖，还能按全站、栏目、详情三级作用域分别生效，改版不用提心吊胆，也不用为每个站各养一套前台。  发布即静态，改一篇文章不用全站重建。内容发布时生成 Astro 静态站，CDN 直达读者。动态后台只在写稿时用，线上跑的是静态文件：打开快，而且线上根本没有动态后台可供攻击，攻击面小得多。构建粒度可以细到站点、栏目、单篇文章，改一篇文章只重建这一篇，不用全站重来。发布成败和耗时都会进铃铛中心告诉你，运维的人不用猜。  AI 写作在编辑器里，不在另一个标签页。很多 CMS 的 AI 是外挂一个聊天框，写一段、复制一段、粘贴回来再排版，切换成本全在用户身上。AI-CMS 的 AI 助手长在编辑器里：起草、改写、润色直接对正文动手，还有上下文感知的快捷操作，覆盖文章、模板和通用对话；增强工具包括文风清理、润色和图片本地化。Provider 可插拔，OpenAI 兼容接口、Anthropic 、MiniMax 、Ollama 本地模型都能接，API Key 只存在自己的服务器上。对同时写几个站的人来说，省的不是某一次写作，是每天几十次切换。  自托管一条命令，内容和 Key 都不出自己的服务器。Docker Compose + nginx 一条命令起全套，4GB 内存、10GB 磁盘就能跑。不按站点数、不按人头付费，不怕平台改价或关停；你的内容、你的 API Key 都留在自己的服务器上。Apache-2.0 开源，v1.0.0 已发布。这对把内容当资产的人是决定性的：工具可以换，数据和账号体系不能被别人捏着。  仓库地址 holyxing/ai-cms ，Apache-2.0 开源。

PolyPixel development for mobile developers, I'm still making PolyPixel, it's getting better and better everyday, I'm working on mobile, please bare me, also I'd appreciate help and contribution on the tool, let's make a 3D tool for mobile together : prime625/godot-vfx-toolkit my phone was broken since September, jus fixed it 3 days back, that's what delayed me cause I don't have a desktop to continue

Bespren Johny Fried Chicken Order Management System. A web app for managing orders of a real business with landing page content management system. Website link: besprenjohnyfriedchicken.pythonanywhere.com GitHub source code (Read LICENSE.md for usage): khianvictorycalderon/BJFC-Order-Management-System t.co

@HaofengC @OpenBot_ we have a watch issue open already openbot#2855

New Laravel package! lmottasin/laravel-http-mailer: Laravel mail over HTTPS: send through the SendGrid or Brevo API when SMTP ports 25/465/587 are blocked on DigitalOcean, AWS, Google Cloud or Hetzner. No code changes needed, only .env changes. lmottasin/laravel-http-mailer t.co

ぬん Strata v0.2.10 に docs を追従 strata-envvars#9

今後機能を拡張するときにセキュリティなど気にした方がいいことを追記 秘密の置き方と公開範囲をdocsに残す github.com/ryo-i/next-nest-started/commit/253c0bc…

Show HN: Elpis – Rust-style ownership for teams of coding agents

@TheWizWiki Es impresionante como estan perfecionando el emu en ps5,es pura fantasia. github.com/Swordpdf/PS5SX2/releases/tag/vk-285-187

DiscImageStudio: burn visible artwork onto CDs, DVDs and Blu-ray discs [Discussion]

Coupling cellular trajectories and RNA kinetics on continuous expression manifolds with ManifoldVelo Zhen Zhou, Jiachen Li, Yuan Liu, Xiaoyong Pan, Hong-Bin Shen arxiv.org [𝚚-𝚋𝚒𝚘.𝚀𝙼] 💬Code: ZHOUZHEN2002/ManifoldVelo t.co

Git-Hub について GitHubとは （ギットハブ） サービス提供元：米 GitHub社（保守対応も含め） ※2018年 マイクロソフトの傘下。 ソフトウェア開発のプラットフォーム。コードのバージョン管理システムにはGitを使用する。 複数人のエンジニアがリモートリポジトリとして活用する他、チーム開発を行うための機能を提供するWEBサービス。 リポジトリとしての機能を持つ他にも、コードレビュー機能やWikiなどのコミュニケーションツールとしての機能を持ち、組織規模を問わず、多くの企業・団体がソフトウェア開発で利用する。 GitHubにソースコードをホスティングすることで複数人のソフトウエア開発者と協働してコードをレビューしたり、プロジェクトを管理しつつ開発を行うことができる。 ・(サイト引用)GitHubの基本機能 ・(サイト引用)GitHubの導入からブランチ作成 ・(サイト引用)GitHub一連の処理の動き [用語] 項目 説明 Issue 課題、修正点 Pull Request Files Changed Pull Requestで表示されるファイルの変更数 リバート バージョニングを一つ一つ戻ること ブランチ 特定の変更点から分岐点までのこと featture 機能開発 リポジトリ バージョン管理によって管理されるファイルと履歴情報を保管する領域 ステージング 作業ディレクトリで編集した変更内容を、次のコミット（履歴の記録）に含めるために一時的に待機・準備させておく場所 バージョン管理基礎（バージョン管理システム） github.com/SMYT-BT/My-initiative/blob/main/Raiset…〇Git（現場での使用率：高）【各端末へ分散型】 バージョン管理システムは大きく「集中型」と「分散型」に分けられる。 ●分散型 Git Hubと連携されているため、サーバーの構築は不要 ファイルの追加や変更の履歴情報を管理することで、過去の変更箇所を確認する、特定時点の内容に戻す、などの「バージョン管理」という作業が可能となる 「分散型」のバージョン管理システムでは個々人のマシン上にリポジトリを作成して開発を行うことができ、現在のチーム開発における主流となっている。 分散型のバージョン管理システムであるGitでは、まず個々人のマシン上にあるリポジトリ上で作業を実施後、作業内容をネットワーク先のサーバー上などにあるリポジトリに集約する流れで開発を進めていく。 この個々人のリポジトリを「ローカルリポジトリ」、集約先となるリポジトリを「リモートリポジトリ」と呼ぶ。 ・基本的な流れ ①ローカルリポジトリにリモートデータの取得 ②ローカルリポジトリでファイル更新を履歴に反映 ③リモートリポジトリでローカルのデータを反映 補足 コメント、承認：レビューの申請と承認 → レビューで承認がないと統合されない 複数の各々のブランチを統合（マージ）するのが主流 ブランチを作成すると、そのコードはブランチを作成する前のコードから分析し、「別の時間を生きる」こととなる。 別の時間を生きたコードは自分だけの経験（変更）を蓄える。 上記プロセスに問題がなければ、親（分岐元）は「マージ（記憶を統合）を要求」する マージは基本リモートリポジトリで実施する ●集中型 SVN【中央集権型】 ※管理するサーバーは自分で立てる バージョンを管理するリポジトリは単一。 リポジトリが壊れても、スナップショットで復旧が可能。 Gitに比べて容易であり、学習コストが低い。 特定の場所にあるリポジトリへの接続が必須となる。 (サイト引用)リポジトリ作成方法 機能 ブランチ保護：GitHubは直接コミット禁止によるブランチの保護（保護されたブランチ/protected branches）を提供する。 GitHubにホストされたリモートレポジトリはgit pushにより更新できる。 →これを許容すると意図しないバグによりpushを受けたブランチが壊れるリスクがある。 GitHubは「指定ブランチへの直接コミット禁止 + チェック通過Pull Requestを介したmerge/rebase許可」という機能を提供することで、ブランチに問題のあるコミットが混入しないことを可能にしている。 一連の流れ | 対応 | 説明 | | git add | インデックスにファイル等を認識させる。[git add .]で全体の変更を更新 | | git commit | リモートからローカルリポジトリにファイル更新の差異を登録する | | git push | ローカルからリモートリポジトリへ差異を連携する | コマンド 項目 説明 git rm -rf {レポジトリ名} git clone 削除 git rm -rf .git git 削除 git ls-files --stage インデックスの中身を見ることができる。 git checkout「ブランチ名」 ブランチへの移動 git branch ブランチ一覧表示 git branch 「ブランチ名」 ブランチ追加 git branch -d 「ブランチ名」 ブランチ削除 git branch -D 「ブランチ名」 ブランチ強制削除 git clone リモートリポジトリをPCに複製する git fetch リモートリポジトリ（GitHubなど）の最新の変更履歴やブランチの情報を、ローカルリポジトリにダウンロードする。自分の手元にあるファイルや現在のブランチは自動で書き換わらない git reset add リモートリポジトリの接続先を登録する git reset HEAD git add の取り消し git remote 登録しているリポジトリの確認 git pull gitリポジトリにある最新のソースコードを取得 git add -u Git が追跡しているファイルの変更・削除をステージングする。新規の未追跡ファイルは追加しない。git push の結果で変更処理ができていないときに実施するとよい git branch --show-current 現在のブランチ名を確認 git branch -m master main ブランチ名を main に変更 [例] $ git remote -v origin github.com/%EF%BD%9E (fetch) origin github.com/%EF%BD%9E (push) ⇒上記は、GitHub上のリポジトリが origin という名前で登録されており、取得と送信の両方に同じURLが設定されていることが分かる。 〇(fetch：取得)リモートからデータを取得するときの接続先。 GitHubにある変更履歴やブランチの情報をローカルに取得します。 〇(push：送信)リモートへデータを送信するときの接続先 ローカルでコミットした変更をGitHubへ送信します。 ローカルで削除した結果を反映させる方法 git add -u # update option git commit -m "deleted some files" git push add → commit → push git push リポジトリ名(※origin) ローカルブランチ名:リモートブランチ名 Q&A GitHub ReadMe とは リポジトリに訪れた人に "このプロジェクトが何なのか" をわかりやすく伝えるための説明書のようなもの。 ユーザー設定確認方法 次のサイトを参考。（参考サイト） Git Cloneの意味 リポジトリのクローンを作成して、コンピュータ上にローカルコピーを作成し、これらの2つの場所で同期することができる。 GitHub.comにリポジトリを作成した場合、それはリモートリポジトリとなる。 リモートリポジトリをそのまま自分のローカル環境（自分のPC上）へ複製（コピー）する機能。 GitHub.com からローカルコンピューターにリポジトリのクローンを作成して、マージの競合の修正、ファイルの追加または削除、より大きなコミットのプッシュを簡単に行うことができる。 リポジトリのクローンを作成する場合は、リポジトリを GitHub.com からローカルマシンにコピーする。 リポジトリをクローンすると、その時点で GitHub.com にあるすべてのリポジトリデータの完全なコピーがプルダウンされる。 これには、プロジェクトのすべてのファイルとフォルダのすべてのバージョンも含まれる。 インデックスがなぜ必要か インデックスとは、リポジトリに保存されている情報とワークツリー（作業している場所）との差（変更箇所）を記録する場所。 その差（変更箇所）だけをリポジトリに保存していく仕組み １年前の GitHub を掘り起こしてみた件 ・端末を変更していたため、Gut用のディレクトリなどはない →つまりローカルリポジトリから作製が必要になった。 ※リポジトリへの接続設定がないか、念のため確認 git remote -v：他のリポジトリへのリモート接続の一覧を表示するコマンド →ローカルの情報？ ★ローカルのリポジトリ名変更方法 qiita.com git clone →リモートリポジトリの環境をローカルにもってくる git reset HEAD：git add の取り消し →念のため、git addの取り消しを実施 git add →変更した内容を加える * Please tell me who you are. Run git config --global user.email "you@example.com" git config --global user.name "Your Name" to set your account's default identity. Omit --global to set the identity only in this repository. →初回扱いであり、必要な情報を以下のように登録していく git config --global user.email "magata.mnm47@outlook.jp" git config --global user.name "SMYT" git commit -a →Commit処理 【エラー】Aborting commit due to empty commit message. →初回のコミット時にメッセージがないため git commit --allow-empty-message -m '' git push origin mian:main →成功 $ git checkout -b lecture02 Switched to a new branch 'lecture02' →ブランチ lecture02 を作成 $ git branch -a lecture02 main remotes/origin/HEAD -> origin/main remotes/origin/main $ git branch -vv lecture02 c2c147a main c2c147a [origin/main] →作成されていることを確認 作成したディレクトリがGitHubに反映されない →ファイルを中に作成しないと反映されないみたい

Codex Review 2026: What OpenAI's Coding Agent Actually Costs (and Where It Loses to Claude Code) OpenAI's Codex isn't a new product announcement — it's been live for a while, bundled into ChatGPT rather than sold on its own. That bundling is exactly why it's easy to underrate: there's no dedicated landing page shouting about it, so a lot of people paying for ChatGPT Plus don't realize they already have a CLI coding agent, a VS Code/Cursor/Windsurf extension, cloud-hosted long-running tasks, and automatic PR review on GitHub, all included. We wrote it up the way we write every tool here — pricing and docs pages checked directly (2026-10-26), cross-referenced against the open-source openai/codex repo, no lab benchmark run, and anything the sources don't state (MCP support, a JetBrains integration) left as unverified instead of guessed. What you actually get: - Plus, $20/month — CLI, IDE extension, web tasks. Usage is capped per 5-hour window, with the exact message allowance varying by which model you're using. - Pro, from $100/month ($100/$200/$500 tiers) — removes the 5-hour cap entirely. - Business, $25/user/month ($20/user annually, 2+ seats) — same per-seat limits as Plus, plus admin controls. - Free and Go — Codex only, through the ChatGPT desktop app, one model, "subject to rollout." - API key — skip the subscription, pay OpenAI's per-token rate instead. The 5-hour window is the detail that matters most if you're deciding between Plus and Pro: OpenAI publishes it as a range per model rather than a fixed number, because the real ceiling depends on task size. Budget for "it varies," not a specific count. Where it actually stands against the competition: - vs. Claude Code — the closest match in the whole market. Same shape: CLI, editor extension, desktop app, web, bundled into a subscription instead of metered separately. Which one wins for you comes down to which subscription you already hold and whose usage limits fit your day better, not a capability gap. - vs. Cursor / Copilot — different category. Codex has no inline autocomplete; plenty of people run it alongside an autocomplete tool rather than instead of one. - vs. Aider / Cline — those are open source and model-agnostic (bring your own API key, any provider). Codex locks you to OpenAI's models but trades per-token billing for flat plan pricing — a real tradeoff, not a strict downgrade. Our take: 4/5. If you already pay for ChatGPT Plus, Codex costs you nothing extra and is worth trying before you add a second agent subscription on top. The honest limitations — no autocomplete, OpenAI-only models, a usage cap that only Pro removes, unverified MCP/JetBrains support — are exactly the things we'd want to know before signing up, not after. Full review with the complete pricing table, strengths/limitations breakdown and "who should use this" guide: devtoolsreview.com We have no affiliate or commission relationship with OpenAI for Codex — checked 2026-10-26, no program exists to disclose.

思考実験として作ってみた winebarrel/terraform-provider-tfcode

@AIMadTech I've mostly been using this persona plugin which has all the stuff I used to use rammy for but it works nicely with Claude and codex: IrrealityLabs/persona-plugin

1) Je vois une vidéo intéressante avec Claude et lui demande de l’analyser via morluto/rea. 2) Je lui demande de mettre en place heygen-com/hyperframes pour répliquer la vidéo et même faire mieux. 3) Je publie. Même pour un néophyte, ça prend quelques minutes et le résultat sera meilleur que ce qu’il propose. À vous de pousser le rendu et de créer vos assets

Check out my recent research about the tld. I shared it on my github: I’m open to feedback and PRs 🤠 jejex0/tld-abuse-intel

Made my own Hyprland shell on CachyOS, where your workspaces turn around a planet ETEA has been my daily desktop on CachyOS for a while now, so it felt right to show it here first-ish. It's a full shell for Hyprland 0.56 (Lua config) on Quickshell: a planet of pages you turn with Super + wheel or a four-finger swipe, an orbit view, worlds instead of themes, desktop widgets, a sky-map launcher, a control centre and a hand-drawn lock screen. On my i5-1335U / Iris Xe laptop it sits around 2.3 GB of RAM for the whole system, with steady frames. Everything it needs is in the repos and the AUR. The install steps are in the README: SDcold/etea-shell If you try it, tell me what breaks!

I built a feed that Claude and ChatGPT can save to, so useful stuff from my chats stops getting lost I'm the maker of Cardlake. I use Claude and ChatGPT daily, but useful things kept getting buried in chats. Links, ideas, reminders, follow-ups. I saved plenty but rarely came back to them. So I built a shared feed for things I actually want to revisit. For example, tell ChatGPT "remind me to send the recap Friday." Later, you can find that card from Claude too. Cardlake also lets you add cards directly, pin, snooze or complete them. Nothing gets saved automatically. It doesn't read your chat history or act as background memory. And it doesn't send push, email or SMS notifications. Try Cardlake: cardlake.app Open-source edition (AGPL): cardleforall/cardlakeforall I'd love honest feedback. What's confusing, and what would make you open this every day?

Motty: SSH, SFTP, port forwarding, VNC/RDP, and AI assistants — all in one native window. 👉 t8y2/motty

#HS2 (#HoneySelect2) の mod plugin 情報：AdvancedLoader キャラや衣装のファイルを読み込むとき、現在のキャラ・衣装の一部だけ上書きする。 対象を細かく指定できるので、類似した衣装ファイルが複数あって、全部「少し更新したい」というときに便利そうだ。 Aktymel/AdvancedLoader

@CristianRus4 Unc someone had already built it 5months ago github.com/Indra55/bezel/releases/tag/v0.1.1

M-WAVE FM-1: Ultimate Unbricking Flowchart Hi all, I come to share a guide to unbrick the M-Wave FM1 via @thehu.cnx & @jinkang_liang on instagram. Quixotic7/MvaveFM1Unbricker Support them and upvote if you like, thanks. Bless you all and hope the info help you

I open-sourced GhostTab 👻: a CLI that lets AI agents (Claude Code, Codex) use your real browser as text, in a hidden window. • ~5–10× fewer tokens than screenshots • Fills whole forms in one call • Safety rails for Submit/Pay Free, MIT, macOS: NadirAliOfficial/ghosttab

I think they namecheck you in AmySynth anyway for your FM Synth work. I can't find it now, but somewhere off shorepine.github.io it talked about the Android work you did. Edit: There is shorepine/alles but I remember a more detailed thing too. I always notice your name because I enjoyed Advogato back in the day and keep using bits of your work elsewhere.

@LangChain @crewAIInc Building multi-agent systems or MCP tool chains? Guardian MCP adds identity + authority boundaries so agents stay scoped and don’t inherit extra privileges across handoffs. npm: @odingard/guardian-mcp GitHub: Odingard/guardian-sdk Open core — useful for production crews/graphs.

Dougie's Balatro - Version 1.0.1 out now! (Check the edit on the end of the post); In this version's fixes, Dougie (u/ Complex_Banana3507) fixed the bug on the corner of the screens in the CRT Shader and my personal biggest problem with this port that was the low frame rates when you progressed into the run; now with these two problems are fixed and also some other optimizations in the bloom filter. With this, now this one is for me the definitive way of playing Balatro on the 360. Please relate any concern or problem on the issue page, not in this post. (edit: a new version is out! 1.0.2; I'm not gonna do an update post everytime of the projects that I follow to not spam the sub, not with updates too closes like this one; but even better overall performance was achieved. The download link is updated too) Download: github.com/dougieprishpreed/balatro360/releases/t…

Show HN: Recursive Human Intelligence: Actionable feedback from your LLM chats Before Recursive Super Intelligence.. a lot of alpha can be squeezed by recursively improve humans.. Dont you think? That's the idea of Recursive Human Intelligence. Practically: it is a desktop that reads Claude Code, Codex, Pi chats, groups them into patterns and gives you specific feedback: touching habits that cost you time. It is bring-your-own LLM; in terms of data privacy it is operating on things you already shared with LLMs. I have gotten some good feedback myself already: on how I wrongly prompt sometimes, so it suggested a course I can take will and a book on MLOps to fill some gaps I miss. I hope it does the same for you! Feel free to contribute :>

做了个给 AI agent 用的 iOS 选库目录，附带一个从零搭项目的 skill 让 agent 新建 iOS 项目时，它选库通常凭记忆，有时会选到已经很久没发版的库。我把常用的 iOS 开源库整理成了一份目录，只放数据，不写主观评价。 - 每个库都列出 star 数、最近一次 Release 、最近一次提交、最低 iOS 版本、Swift tools 版本，由 GitHub Actions 每周自动刷新 - 首选按固定规则算出：超过两年没发版的、已归档的排除，剩下的按 star 排序。后端、付费、数据分析这类服务商 SDK 不排名 - 每个分类都附上能做同样事情的 Apple 原生 API 和最低系统版本 几个两年以上没发版的库：Charts （ 28k ★，最后一次发版在 2024-02 ）、Moya （ 15k ★，2022-08 ）、KeychainAccess （ 8.3k ★，2021-03 ）。 另外附带一个 ios-starter skill 。agent 会先问你的需求，再按上面的规则选库，然后用 XcodeGen 、Tuist 、MobileBuildMCP 或 Xcode MCP 四种方式之一建项目，最后编译并在模拟器里启动。这四种方式我都在 Xcode 27.2 上完整跑过。 npx skills add jeffy-w/awesome-ios-for-agents@ios-starter -g -y 目前收录了 55 个分类、91 个库和工具，有中文 README 。觉得漏了哪个库，或者分类不合理，欢迎提 issue 或 PR 。 jeffy-w/awesome-ios-for-agents

This shows your AI agents what they are building. On a map. Show HN: Archwhere: • Work alongside AI coding agents • A live architecture map • See which part of the product each agent changes Architects finally get agent visibility. tapaskar/archwhere

WoW Forever character & alt planner (Excel). Looking for feedback >Upfront: I used AI to help me research and build this. The result is NOT an AI tool, though. It’s a plain Excel file with no AI inside, no add-ins and no macros. Hi fellow enthusiasts, I had some spare tokens left and used them to go through the information available so far, mainly wowhead and blueposts posts, plus a few other sources. I also added my own knowledge and, admittedly, some personal preferences, and turned it all into a character planning workbook that I’d like to share with you. Why I built it I love the journey more then the endgame. That is why in the long run I want to play every class. Not all at once, but over time I’ll build up a whole army of alts, like I always did on every release :) As it can become challenging to keep track I want to make sure to keep an overview and that: - all crafting professions are covered, - I have the gathering professions to supply them - all tracked per ruleset and faction The overall goal is to find good combinations: faction, race, class, spec, professions, and which classes team up well. Who it’s for You can plan: - just your own characters, - you plus a buddy, - or a "butler" character, which some content creators talk about right now: a second character on a second battle.net account that you play alongside your main (multiboxing with yourself with manual operations, basically. Compliant with ToS). It’s about high-level decisions, not min-maxing. So there’s no gear or BiS planning in here, so mostly just the big "one time" choices. What’s inside - Roster: one row per character, with dropdowns that only offer valid race, class, spec and profession combinations. Checks flag problems: wrong faction, a crafter with no gatherer to supply it. - Coverage: which professions, classes and races you already have, counted per faction. - Best Combos: Wowhead’s class guides boiled down to the best race and professions per spec and faction, plus the tier lists. - Duo Matrix / Duo Combos: which two classes level well together, and why. Useful on its own, too. - Ref tabs: all racials (Forever values after the tuning passes), the race/class grid including Skyborne, and professions. Caveats: the data was checked against Blizzard’s launch dev notes and wowhead today on November 3. Some wowhead class guides have not been updated since recent changes. Requires Excel 2021 or Microsoft 365. What I’m asking for: constructive feedback. I’d love to hear: - Errors: anything wrong or outdated. - Gaps: anything crucial that’s missing. - Improvements: how to make it easier to use or more useful. - Features: ideas for what it should do next. I am aware that many of the current information is likely to be changed or at least adjusted. I’ll keep improving it and share updated versions here (via same github link). Download: github.com/wowforeverplanner/wowforeverplanner/re… Have fun, and see you in Azeroth!

🚀 Trending on GitHub Hex by anomalyco ⭐ 4,218 (1,134 today) 🧠 Hold a key, speak, and it types for you, transcribed fully on your own device, nothing sent to the cloud. anomalyco/hex #GenAI #OpenSource #AI

A whole Wayland desktop shell in QML: planet page transitions, live window copies, and idle CPU down from 15% to 3% ETEA is a desktop shell for Hyprland, written entirely in QML on top of Quickshell. Bar, widgets, launcher, notifications, lock screen, and a page switcher where your workspaces turn around a planet. A few things I learned keeping it smooth on a laptop iGPU (Intel Iris Xe): - An invisible item that still changes still costs a redraw of the whole layer. The big scenes live in Loaders and only exist while they play. - Every animated decoration shares one 15 fps clock that items join and leave, instead of each running its own timer. - During a page turn, windows are live ScreencopyViews over a still photo, blurred with MultiEffect where the real app is translucent. - Sky, ground and card textures are small GLSL shaders compiled with qsb. - Sensors are read with FileView on sysfs rather than by starting processes. Together that took idle CPU from about 15% to about 3%, and the whole system sits around 2.3 GB of RAM. Code (GPL-3.0): SDcold/etea-shell Happy to talk about any of it, or to hear how you'd have done it.

📥更新に失敗する方はこちらのv0.5.0.94をお試しください！ ご利用中のNexus Arkを終了しフォルダも閉じて、新しいZIPを別の場所に展開。その中の「ImportUserData.bat」から今までのNexus Arkフォルダを選択。引継ぎ終了後、新版フォルダのStart.batから起動。 github.com/kenomendako/Nexus-Ark/releases/tag/v0.… t.co

Agents registered on Base or Ethereum can now have a Bursar endowment on Robinhood Chain. An attestor records the owner, who must accept before any draw reaches the chosen spending mandate. The funding stays on Robinhood Chain. github.com/bursar-world/bursar/commit/6f40b0ceb2d… github.com/bursar-world/bursar/commit/3c3864e6282… bursar.world

3rd career-ops PR merged🙌 a dns server that never answers used to hang a plugin call. now it stops after 5 seconds and tells you why exa plugin in the registry 10 MB cap on plugin responses this one thanks @santifer for reviewing all three career-ops#6404

@LenSeaside @UnoPlatform You can check project here wieslawsoltes/UnoPixelMap

Inside Angular’s Change Detection Zones: How UI Updates Are Scheduled and Batched You’ve just fixed a bug where your Angular component’s UI didn’t update right after an HTTP request. You refresh, hoping the input reflects the latest data immediately. But nope, the UI lags, or requires a second click, or only updates after you interact again. What gives? I’ve been there. It feels like Angular’s change detection is on its own schedule, ignoring your updates. Tracking down the root cause led me straight into Angular’s NgZone and how it intercepts asynchronous operations to schedule UI updates. The Familiar Developer Frustration: UI Doesn’t Update When You Think It Should Imagine this: you call a service that fetches data with HttpClient. You subscribe, set a component property in the callback, and expect Angular to refresh the template. this.http.get('/api/data').subscribe(data => { this.user = data; // Expect UI to update immediately }); But sometimes the UI stays stale. Or, if you’re using third-party libraries or custom async code, change detection doesn’t run automatically. You might resort to triggering it manually with ChangeDetectorRef.detectChanges() or NgZone.run(). That’s a red flag something subtle is happening under the hood. What Is NgZone and Why Does Angular Use It? Angular wraps your app’s execution inside an NgZone, an Angular-specific zone based on the zone.js library. Zones monkey-patch async browser APIs like setTimeout, promises, events, and XHR. This lets Angular know when async work starts and finishes. Why does it matter? Because Angular’s change detection is not running constantly or on every microtask. Instead, Angular waits for these async activities to complete and then triggers change detection once per turn of the event loop. This batching avoids excessive UI work. In short, NgZone is Angular’s way to track async boundaries and schedule UI updates efficiently. How NgZone Intercepts Asynchronous Operations Behind the scenes, zone.js patches these APIs: - Timers: setTimeout, setInterval - Promises: .then() callbacks - DOM events: addEventListener - XHR and fetch Each async callback runs inside the NgZone. When an async task starts, NgZone increments an internal counter. When it completes, it decrements it. When the counter hits zero, NgZone knows the async turn is over and calls Angular’s change detection. This is why Angular’s change detection doesn’t run immediately at every microtask or callback, but batches updates until the async “turn” is done. What Happens When You Run Code Outside NgZone? Sometimes, to optimize performance, you or a library might run code outside Angular’s zone using NgZone.runOutsideAngular(). This means Angular doesn’t intercept those async callbacks. this.ngZone.runOutsideAngular(() => { setTimeout(() => { // Angular change detection won't run after this timer this.user = newUser; }, 1000); }); If you update component state inside such callbacks without re-entering Angular’s zone, your UI won’t update automatically. That’s a common source of confusion. The fix: wrap updates inside NgZone.run(), or trigger change detection manually. The Tradeoff: Performance vs Predictability NgZone’s interception is powerful but comes at a cost. Monkey-patching async APIs adds overhead, which can affect performance in large apps or complex async flows. Because Angular batches change detection after all async tasks finish, UI updates might feel delayed if you have chained or long-running async operations. Some experiments and libraries aim to avoid zones entirely (like Angular Signals or manual change detection strategies) for more control and performance, but zones remain the default for good developer ergonomics. Examples: When Change Detection Runs Example 1: Simple Promise Promise.resolve().then(() => { this.count++; // Angular will detect changes after this microtask finishes }); Angular patches promise .then(), so after the microtask queue empties, change detection runs. Example 2: setTimeout setTimeout(() => { this.count++; // Angular detects changes after timer callback }, 100); NgZone tracks timers similarly, triggering change detection once the callback completes. Example 3: Event Listener document.getElementById('btn').addEventListener('click', () => { this.count++; // UI updates because event listener runs inside NgZone }); Events patched by zone.js trigger change detection after handlers. Example 4: Outside Angular Zone this.ngZone.runOutsideAngular(() => { setTimeout(() => { this.count++; // UI stays stale because this runs outside Angular zone }, 100); }); No automatic change detection unless you re-enter Angular zone or detect changes manually. Debugging Unexpected UI Update Timings If your UI doesn’t update as expected: - Check whether your async code runs inside NgZone. - Use console.log(NgZone.isInAngularZone()) inside callbacks to verify. - Try forcing change detection temporarily with ChangeDetectorRef.detectChanges() to isolate the problem. - Beware of third-party libraries that operate outside Angular zones. When Change Detection Might Feel Too Aggressive Because NgZone triggers detection after every async turn, sometimes Angular runs more often than you want, especially with frequent timers or event-heavy components. You can optimize with: - ChangeDetectionStrategy.OnPush to reduce detection scope - Running complex code outside Angular zones if UI updates aren’t needed every time - Using signals or other reactive primitives for fine-grained updates Wrapping Up NgZone is Angular’s secret sauce for knowing when async work finishes so it can batch and schedule change detection. It patches your async browser APIs, tracks when tasks start and finish, and triggers UI updates accordingly. This makes Angular apps feel reactive without you having to call detectChanges() everywhere. But it can also surprise you when third-party async code runs outside Angular’s zone, or when UI updates lag behind your expectations. Understanding NgZone’s role helps you debug those subtle UI update glitches and write more predictable, performant Angular apps. --- Helpful learning resources - MDN Web Docs - web.dev performance guidance - OpenAI developer resources Originally published at Under The Hood. Get the next deep dive in your inbox: subscribe to Under The Hood.

If no one continue $PMUP, I'll just continue it on 5fmDXRXNm9gM9QP4CJk2QEuhqCHoTXc3pL4WB47opump $PUMP lol 😂😂🤣🙏 AlleyBo55/pumpfun-source

@CallanHardin @UnoPlatform You can check project here wieslawsoltes/UnoPixelMap

One of my recent portfolio projects took me beyond dashboards and into the work that makes analysis reliable. Cleaned 20,000 Flipkart product records using Excel & Power Query, uncovering duplicates, inconsistencies in formatting, missing values and more github.com/Freddiecoder99/Data-Analysis/tree/main… t.co

GitHub actionsのCIはテストのみだったのでデプロイは別であることを追記 CIとデプロイの関係をdocsに残す github.com/ryo-i/next-nest-started/commit/534b296…

If you need keyboard, mouse and gamepad input together, I also maintain the combined Interceptor edition: TitanCSA/Interceptor Use its matching driver/SDK. Hardware support varies. MIT SDK; private driver, admin, 1h/kernel-boot demo; full pricing by agreement.