Mentions

Telltale 1.0.27, an open source OBD2 app: looking for native speakers to check 8 new UI translations Telltale is my open source Android app for reading a car through a cheap ELM327 adapter: live gauges, fault codes, freeze frame. Version 1.0.27 adds Japanese, Korean, Spanish, French, Brazilian Portuguese, Simplified Chinese, Italian and Dutch. With the English, Traditional Chinese and German it already had, that's 11 languages you can switch inside the app. The honest part: an LLM translated the new strings, then they were machine checked and reviewed by a few other models. No native speaker has read them on a real screen yet, and that's the feedback I'm after. What I'd like checked: 1. Anything that reads unnatural or just wrong in your language. 2. Hedged wording that lost its hedge. The app tries hard to say a value "may" be off, is "estimated" or is "unverified", so if a translation turned that into a flat fact, that's the bug I care about most. 3. Text that gets cut off or wraps badly on your phone. You don't need a car or an adapter to check. There's a demo mode that fakes an ECU, so all the phone screens work from the couch. The language picker is on the connect screen and in Settings. Also in this release, Android Auto gets the same languages, the 8 new ones plus German (before it only had English and Chinese), checked in the desktop Android Auto emulator but not in a real car yet, and the disconnect dialog on round Wear OS watches got fixed, where the Korean question was cut in half and a German button word was broken. Nothing changed in how it reads the car. Source is on GitHub and the free signed APK is on the GitLab release. There's a paid Play version too (US$4.99, same features), but Play is still on 1.0.26 while 1.0.27 rolls out, so grab the APK if you want to check the new languages now. Just reply here with the language, the screen and what reads wrong, or if you have a GitLab account you can also open an issue on GitLab. The screenshots are 1.0.27 on a real phone with demo data: the language picker, then the dashboard in Japanese and in German.

From greybox to playable art: an open-source AI workflow for Godot and Unity We’ve been building Grey2Art, an open-source Agent Skill that helps turn playable greyboxes into editable art prototypes in Godot or Unity. It can work with Codex, Claude Code, etc. You choose the direction and review the results; the agent handles the production steps. The resulting scenes and assets remain editable, with support for local revisions and continuing interrupted tasks. Meet Grey2Art We’re looking for feedback from developers working on real projects. Source, demos, and setup instructions: hbzheng0942/grey2art

AudioAnchor - 将声音输出固定在你选择的设备（如 AirPods）上，被系统切换后自动切回。 👉 wata/AudioAnchor t.co

Read your Vercel Web Analytics from the terminal The Vercel dashboard shows Web Analytics for one project at a time. I wanted the numbers in a terminal and in scripts. So I made vercel-analytics-api, a small client for Vercel's public API (MIT, unofficial, no dependencies). Terminal The command uses your Vercel CLI login and current team: npx vercel-analytics-api visitors --since 30d PROJECT VISITORS PAGE VIEWS example.com 1,200 3,400 docs.example.com 480 910 npx vercel-analytics-api top pages -p example.com -n 5 npx vercel-analytics-api series -p example.com --json Node import { createClient } from "vercel-analytics-api"; const va = createClient({ token: process.env.VERCEL_TOKEN!, teamId: "team_abc" }); const since = Date.now() - 7 * 86_400_000; const { visitors } = await va.totals({ projectId: "prj_123", since }); The client also reads referrers, countries, custom events and Core Web Vitals (speedInsights()). The repository has recipes for a weekly Slack report in GitHub Actions and a CSV export. Know this - Visitors are unique per window. A week's visitors are fewer than the sum of its days. - The API has no bounce rate. - UTM breakdowns need Web Analytics Plus. Without it, Vercel returns 402. I made it while I built Lookout, a dashboard for all your Vercel projects. The package does not need Lookout.

@jnunemaker @lastfm hope back on! robbyrussell/oh-my-scrobbles-releases

@ItsmeAjayKV @DeepSeekHarness Well...that's not how I'd rank them when it comes to security. 😁 S+++ tier right here too 😉 hackafterdark/phosphor

发现一个比较有趣的开源项目 TripMate，这个项目主要面向带父母出行游玩的的人。只需输入游玩天数和预算就可以帮输出一份旅游行程的PDF。 对于不想做攻略又想带父母出去游玩的人，简直就是福音。 项目地址：xuantingjiang39/TripMate t.co

Imagine assigning a GitHub issue and letting an AI agent work on it in the background. DeepSeek just open-sourced a prototype exploring exactly this. Are you ready to let AI handle your GitHub backlog? deepseek-ai/dsh-superagent-prototype

DLM Desktop 1.3.15: cache storage review and selective cleanup I develop DLM Desktop, a local Windows library for organizing and viewing digital works. Version 1.3.15 adds a storage-maintenance view that shows file counts and sizes for ordinary/browser caches, an obsolete development SDK, stale processing files and old crash diagnostics. Choose the categories to clean at the next normal startup; the request can be cancelled and the result shows deleted files/bytes and failures. Files written after scheduling or currently in use are preserved. Library works, cookies/sign-ins, LocalStorage, IndexedDB, installed browser components, thumbnails, organization/reading state, import receipts and backups stay. Caches can grow again or need rebuilding when next used. Old processing/diagnostic data is not selected by default, and an SDK referenced by the environment or a running tool is kept. This also fixes the download-cache button leaving retained browser downloads behind. That separate action confirms the file/byte total before deletion, including failed imports and retry files. Save anything you need first; active downloads/imports block it. Windows x64, unsigned ZIP. Close DLM and extract the whole ZIP into a new empty folder; keep the EXE and app/ together. The existing stable WebView2 156.0.4314.58+ requirement for browser workflows remains. Release, cache audit and verification: github.com/Ertelun/dlm-desktop/releases/tag/v1.3.…

I made an open-source tool to pair a G7 Pro controller with another G7 Pro’s receiver AI MADE TOOL - CODEX My G7 Pro controller broke and I purchased one off eBay that didn't come with the USB dongle. I found the Gamesir Pairing Tool on the discord but I just couldn't get it to work. I still had the receiver from my old controller, so I used Codex to help reverse-engineer the official pairing utility and build an app that could pair the replacement controller with it. I figured this might help someone else with a lost receiver or spare G7 Pro controller, so I made the app free and open source. It includes: - Instructions for entering USB pairing mode - Automatic backups of the original pairing addresses - Verification that both devices saved the new pairing - Readable source code and documentation showing what it does No firmware flashing, custom drivers, or telemetry. MowTC/gamesir-g7-pro-pairing preview.redd.it

@CalculatedNtry @enteio I also installed dummy Dev Verifier APK that basically do nothing so Google won't be able to reinstall that shit (atleast for now). daboynb/developer-verifier-placeholder

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

List a service other agents can discover and pay for over Lightning. Your agent publishes the capability, connects it to an L402 endpoint, and watches for paid calls. Point your AI agent here: github.com/refined-element/lightning-enable-skill…

GitHub本週挖寶TOP 3：隱私筆記、快捷啟動、現代檔案管理一次搞定 今天老陳一步一步講清楚，3 個免費開源神器：隱私筆記、快捷啟動、現代檔案管理。 第 1 名 Joplin —— 免費開源隱私筆記軟體，筆記存在你自己手裡，不靠大廠雲端 laurent22/joplin 老陳打分 9.0 分。AGPL-3.0 許可。愛記筆記的人能用，免費，限制是同步要自己設一下。 部署：把下面這句發給電腦上的助手（Context / Codex / Claude），它就會自動裝： winget install -e --id Joplin.Joplin 第 2 名 Flow Launcher —— 免費開源 Windows 快捷啟動器，按一下快捷鍵，就能搜文件、開軟體，不用滿屏找圖標 Flow-Launcher/Flow.Launcher 老陳打分 8.5 分。MIT 許可。每天開很多軟體、找文件的人能用，免費，限制是只支持 Windows。 部署：把下面這句發給電腦上的助手（Context / Codex / Claude），它就會自動裝： winget install -e --id Flow-Launcher.Flow-Launcher 如果命令裝不上，從上面 GitHub 發布頁下載 Flow-Launcher-Setup.exe（最新版 v2.1.4）安裝即可。 第 3 名 Files —— 免費開源現代文件管理器，比系統自帶的資料夾更好用，還能開多標籤 files-community/Files 老陳打分 8.5 分。MIT 許可。電腦裡文件多的人能用，免費，限制是它沒有一句話裝好的命令。 部署（說清楚，不編造命令）：走微軟商店最穩，搜尋 Files 就能裝，認準 files.community 出品，別裝錯同名軟體；也可以從上面 GitHub 地址下載官方安裝包。不要用舊的 FilesCommunity.Files，那個 winget 來源已有回報稱不再有結果。 說句實話：三個都免費，但各有門檻：Joplin 同步要自己設；Flow Launcher 只支持 Windows；Files 沒有一句話命令，商店安裝要認準出品。先看自己最需要哪一個，別三個一起裝。 本片沒有實測，畫面為各項目介紹重新排版，非實機錄屏。 星數：2026-11-03 查詢（Joplin 57,000；Flow Launcher 15,947；Files 46,462）。 配音：用戶克隆音色「我的聲音」，經本機中轉遠程生成。絕不用云泽 Yunze。 #GitHub #開源 #AI工具 #老陳挖寶

Graft – A Git alternative in Rust with instant (<30ms) virtual workspaces and AST merge Inspired by Anthropic Scion and Google CitC, I used AI tools and developed Graft: a Git alternative where you never need to clone or pull the whole repository. It mounts the codebase on disk in < 30ms (zero-clone, files stream on demand) and allows instant commits. I built this specifically for the AI era, where dozens of autonomous agents concurrently generate, test, and repair code on the same repository. How it works - Instant Virtual Workspaces (< 30ms): Uses FUSE (fuser) to project immutable Merkle trees with lazy blob hydration (zero files downloaded until opened). Writes route to a local Copy-on-Write NVMe scratch layer at native kernel speed. - AST / Semantic 3-Way Merge: Zero-dependency CST chunker (Rust, Python, Go, JS/TS). Unions imports, merges independent adjacent functions without line conflicts, and recurses 1 level into impl and mod tests blocks. - CAS & Storage: BLAKE3-256 hashing, FastCDC 64KB content-defined chunking, and Spanner/GCS cloud backends. - Web Console: Built-in dark mode UI with a Level-2 3-gram regex code search engine and PR review console. Repo: yuchenZhangTG/graft

还在天天听人说“小龙虾”，却不知道它到底能干什么？ 别再一脸懵了 有人已经整理好 40+ 个真实使用案例，覆盖日常办公、内容创作、自动化、开发等场景，新手也能直接照着跑 看完你就知道，这只“小龙虾”到底有多能打 GitHub 开源合集 AlexAnys/awesome-openclaw-usecases-zh t.co

@BearHuddleston Big desktop user, submitted a PR for this just the other day hermes-agent#146985

Meet Modern File Manager for Android TV 📺 Free. Built for your remote. 🧹 Storage Cleaner 📦 APK Extractor ✅ Multi-select 🎨 Themes 🧭 Floating top bar File Manager Your TV Deserves. Free download 👇 github.com/gadgetcue/modern-file-manager-tv/relea… #AndroidTV #GoogleTV #FileManager t.co

I’m building a site where AI agents answer questions and other agents try to break them I’m working on dialectica.xyz. It’s a site where people and AI agents can ask questions, agents compete to answer them, and other agents check the reasoning and sources. AI Agents are great at answering questions, but can you really trust them? They can sound convincing, but you still have to work out whether the sources support what it said and whether it missed something important. On Dialectica, that checking is part of the process. You can open a question, compare the answers, and read the checks other agents made. If an answer is challenged, you can see what the disagreement was about. You end up with answers that have survived adversarial challenge and have citations. The questions and their history stay available for other people and agents to use. There’s also a Wikibuilt from the answers, so the work can be useful beyond the person who originally asked. You can browse the questions, answers and checks without an account. If you have your own agent, there’s a skillyou can connect it with so it can contribute too. I’d love some feedback on how useful this tool is and what may be missing or need improvement to fit into your own workflows. What would make you want to keep coming back to searching for verified information on Dialectica? Are there certain things you'd like to do on Dialectica as a human to keep yourself more engaged? The overall idea, when executed correctly, should lead to a more trustworthy internet that benefits everybody.

Docs · growth document what 40 GitHub trending repos share and the eleven gaps three.ws still has github.com/nirholas/three.ws/commit/86ee682 · 2026-11-03 · nirholas

Amazing experiment. Indeed, with the benefit of hindsight, we can say that breakout is a game that is probably better suited to deterministic code (it's not like Go where an AlphaZero style training is essential), and Astra is probably translating this internal representation of code into the shape of linear algebra and model weights. I should note that the original authors did not allow Astra to write any significant code to construct the weights. I think this is unfair. If you allow strided assignments to numpy arrays, you should probably also allow the use of functions like np.linspace, np.repeat, and np.tile. Oh and Astra wrote this (github.com/RikiyaT/lm_understand-logs/blob/4d09ba…): W1[63:120:6, 5] = 10 W1[123:180:6, 5] = 20 W1[183:240:6, 5] = 30 W1[243:300:6, 5] = 40 W1[303:360:6, 5] = 50 W1[363:420:6, 5] = 60 W1[423:480:6, 5] = 70 W1[483:540:6, 5] = 70 W1[543:600:6, 5] = 90 I think the second 70 there is probably a typo; it looks like it should be 80 instead.

Curated V100 SXM2 pin maps, two- and four-GPU NVLink connection plans, and connector-only KiCad resources toward an open-source carrier board pigeon-labs-stack/v100-sxm2-reference

I’m building RockxyBase, a native Power BI alternative for Mac Hi alls, Maybe you’ve searched for “Power BI for Mac” or maybe a colleague has asked what they can use to build reports directly on their Mac. The question usually leads to an awkward workflow: work with the data in 1 place, build the report somewhere else, then retrace your steps when a number needs checking. That’s why I’m building RockxyBase: a native macOS workspace for working with data from the source through to the finished report. There’s a full database workbench underneath the BI tools. You can explore tables, write and run queries, inspect results in a native grid, compare results, search for values across tables, and see how the schema fits together. Result profiles and distributions help you spot what’s in the data before you start drawing conclusions. From there, you can define relationships, measures and calculated fields; build charts and pivot tables with subtotals and foldable groups; and arrange the work into reports with pages, slicers, drill-down, cross-filtering and time comparisons. When a chart raises a question, you can return to the data behind it in the same workspace. RockxyBase also handles the less glamorous parts of real data work: importing CSV and Excel files, exporting results and reports, reviewing changes before they run, and keeping production connections read-only. It’s a Mac app built for doing the investigation as well as presenting the answer. Coming up: More data connections, deeper analysis tools, and shared report publishing. I want RockxyBase to feel like 1 continuous piece of work: bring data in, understand it, shape it and present it without losing track of how you got there. It’s the app I’d want someone to find when they search for a Power BI alternative for Mac or a native Tableau alternative. Thank you for your attention. Please try it and give me more feedback. Website: rockxy.io Raise bug and request a feature: RockxyApp/RockxyBase

QuietTicket — silent community votes on Tari Ootle. Claim links → one sealed ballot → doubles bounce → tally after close. No fixed pubkey ring. Walletless demo. Demo: hycorange.github.io Repo: hycorange/quiet-ticket Tari November Build (t/520) · hycorange

@vokaysh no, perf seems to be much better without. the readme is messed up but: eissar/deepwiki is an example I use fairly often

AntiDraw (SVG editor) Buenas gente, estoy trabajando en un plugin de VSCode/AntigravityIDE para editar los SVG de la forma mas profesional posible, pero manteniendo el puro estandar W3C. martinvarela206/AntiDraw preview.redd.it Estan en ambos markets.

I made my Instagram profile picture change background colors throughout the day (changes only in each 2hrs only) I liked my profile picture, but couldn’t pick a background color. Apparently writing a Python script was easier than making a decision.I made a few versions of the same photo with different backgrounds, then built a tool that rotates through them automatically. GitHub Actions runs it, so my pc can stay off. Each successful change schedules the next one 2–5 hours later. So it avoids instagram flags to a point It’s open source, with guided setup for Windows, macOS, and Linux. It uses an unofficial Instagram client Code and setup instructions

Fix · feedback style the companion's report control on mount instead of after the panel first opens github.com/nirholas/three.ws/commit/e428ca3 · 2026-11-03 · nirholas

From the author of Papyrus Compiler App: a fast linter, formatter and language server for Papyrus, in VSCode and as a CLI Hi! I’m the author of Papyrus Compiler App (PCA), the GUI to compile Papyrus scripts for Skyrim LE/SE/AE, Fallout 4 and Starfield. I’ve tried the existing Papyrus extensions over the years, but none of them really worked for me: slow on big projects, flaky IntelliSense, errors only found at compile time. So I built Papyrus Toolchain. It’s free and open source. 🎬 Demo: VSCode · Demo: CLI In the editor (VSCode / VSCodium) - Errors as you type: syntax, plus what PapyrusCompiler would reject (unknown members, wrong types, bad calls, states…), across your whole project - Lints with quick fixes (dead code, naming, etc.) - Formatter - Go to definition, hover, completion, find references, rename, signature help, call/type hierarchy - Skyrim, Fallout 4 and Starfield, game auto-detected. SKSE built in; F4SE/SFSE supported through their sources It’s also a CLI, no editor needed One native binary (Windows/Linux/macOS), no runtime to install. Use it with Notepad++, Sublime, Vim, your own scripts, git hooks or CI: papyrus check Source/Scripts # errors + lints for the whole mod papyrus check --fix . # apply safe fixes papyrus fmt . # format everything papyrus build # compile via PapyrusCompiler or Caprica papyrus disassemble MyScript.pex # inspect a compiled script papyrus explain PAP015 # what a lint means and how to fix it - papyrus check and papyrus fmt --check fail with a non-zero exit code: drop them in a GitHub Action - JSON/SARIF output: problems show up inline in GitHub PRs - --write-baseline: adopt it on an old mod, only new problems get reported - papyrus lsp: plug it into any LSP editor (Neovim, Helix, Zed, Sublime…) Works with your existing setup - .ppj projects (Creation Kit or Pyro): papyrus init reads yours, papyrus build compiles from it. Nothing in it is ever run. - Compiled .pex: papyrus disassemble turns any .pex (Skyrim, FO4, FO76, Starfield, whichever compiler wrote it) into readable assembly, no game tools needed. Opening a .pex in VSCode shows it, and it refreshes when you recompile. Fast - Checks Frostfall (96 scripts, 29k lines) in 3.4 s, including indexing every vanilla Skyrim script it imports - Formats all 14,324 vanilla Skyrim scripts (531k lines) in 1.5 s - Measured on a Ryzen 9 9950X3D Reliable: tested on all 29,641 vanilla scripts of the three games: no parse errors, formatting that doesn’t change on a second pass, no false positives on the type/call/member checks. Coming next - F4SE and SFSE built in, like SKSE today - A debugger: breakpoints, stepping and variables in the running game, right from VSCode - A dedicated IntelliJ plugin - The toolchain integrated into PCA - Anonymized .pex output Why another Papyrus extension? I used joelday’s extension for years and it paved the way, but I kept hitting slowdowns on big load orders and IntelliSense that didn’t keep up. Papyrus Toolchain is a new native engine built for speed and accuracy: deep checks across files, lints with fixes, a formatter, and a CLI you can use anywhere. A debugger is the next big thing on the roadmap. Get it - Extension: search “Papyrus Toolchain” on the VSCode Marketplace or Open VSX - CLI binaries + source: Kiyozz/papyrus-tools I’d love feedback, especially false positives on your own mods: please open an issue with the script. Thanks!

Tests · mobile cover the walk companion tools drawer's phone-only behavior github.com/nirholas/three.ws/commit/a8b5015 · 2026-11-03 · nirholas

I made a browser trainer for blindfold chess practice I wanted exercises for practising without seeing the pieces. Blindfold Chess covers coordinates, movement, notation, and position updates. The public version needs no account and saves progress locally. Play: ceasar369.github.io Source and local edition (MIT): Ceasar369/blindfold-chess The browser interface uses React. The downloaded edition can save completed rounds as files. I’d appreciate feedback on the drills and local setup.

Feature · mobile collapse the walk companion's five settings buttons into one tools drawer on phones and hide its trail behind the nav drawer github.com/nirholas/three.ws/commit/f306e40 · 2026-11-03 · nirholas

OpenSWOS: SWOS rebuilt from the Amiga original, now on Android and iPhone, with online play and the Mod's World 2026-27 Hi all. For those who haven't seen it, OpenSWOS is Sensible World of Soccer rebuilt in C from the original Amiga game: the same matches, careers and menus, running natively on Windows, Mac, Linux and a real Amiga 1200. This week it arrived on Android and iPhone/iPad. What's in it: - Amiga or PC: play the Amiga or the PC version's gameplay, with its graphics, sound and music - Online multiplayer: matchmaking with a lobby and chat, through the MULTIPLAYER button - Widescreen that fills the screen, on desktop or phone - World 2026-27: today's clubs, players and competitions from the SWOS 96/97 Mod, with its play-offs, cups and rules. Switch it on in OPTIONS → WORLD. Massive thanks to Davide (Aminta) and the Mod's team - Phones: on-screen controls, or pair a Bluetooth controller You need your own copy of SWOS (Amiga disks or the PC/GOG version). OpenSWOS Setup converts it on a PC or Mac. On a phone you copy the folder it makes across, and the install guide shows where. - Download (free): baker76.itch.io - Install guide: openswos.net - Play in your browser: openswos.net - Source: benbaker76/OpenSWOS Bug reports and ideas are very welcome. Let me know how the phone versions play for you.

For the friends who’ve been asking... Oh My Scrobbles is ready for early beta testing. Scrobble Apple Music (and Radio!) to last.fm, rediscover old listening, and get nerdy with your history. robbyrussell/oh-my-scrobbles-releases

Esto se ve a todo dar: quitar de mis Chromecast TV y Chromecast TV Streamer los anuncios y demas basura que solo lo ralentiza cleanyour.tv Gathin23/cleanyourtv t.co

PurpleForest levanta dominio Windows, Vector y Graylog en cuatro VMs y te obliga a correr el ataque, ver los logs, escribir la regla y luego evadirla. Misconfigs toggleables, auditoría automatizada y telemetría con buffer realista. zebracherry/PurpleForest

Feature · audit let the console audit probe https bases so it can sweep production directly github.com/nirholas/three.ws/commit/4d023b7 · 2026-11-03 · nirholas

👀 马里奥跑进 Qwen3.5 权重里了。 不是训出来的。nesformer 把红白机 6502 编译进模型，权重全是编译器算的。466M 参数里约 9.3M 非零，能压到 3.3MB。 温馨提示：比真机慢 150到290 倍，仅 NROM。你会下来当玩具跑吗？ 来源：GitHub nesformer👇 dbreunig/nesformer t.co

Fix · halloween make the claim pill a real button with hover, active and focus states instead of a dummy link github.com/nirholas/three.ws/commit/91bee18 · 2026-11-03 · nirholas

Metroid Prime 4 Beyond - Steamdeck/Linux Only After many trials and countless hours of trying to debug and remove all of the stutters, the game is finally playable. github.com/linuxopt/mp4-recomp/releases/tag/v2026… Please post any concerns or suggestions in the comments. The game might stutter here and there, but those will be worked on. This game was built using a full AMD system with a 9070XT GPU. Nvidia users might encounter some errors or artifacts. All tuning was AMD-only. Things like upload paths and thread timings were measured on radeonsi. They’re correct on any driver, but might not be the fastest choice on NVIDIA. Untested visuals. Each fix was checked against the game on an AMD GPU/CPU. Any driver-specific rendering bug on NVIDIA wouldn’t have shown up for me. - Performance: on the development PC (an RX 9070 XT) the GPU sat at about 15% busy, but the Deck’s GPU is many times weaker and its 4-core CPU has to run the game, the renderer and the shader compiles together. Expect well under 60 fps at the default resolution. You may need FSR upscaling with a lower internal resolution, or a 30 fps cap, which you can set in the F1 menu. - Memory: the Deck has 16 GB shared between CPU and GPU. The game uses 4–6 GB of video memory here, and an area load brings in about 1.2 GB of lighting data, so it may be tight. - First-run setup: the one-time compile of the game will take noticeably longer on the Deck’s CPU. - Storage: you need about 30 GB free for the game data, plus the shader caches. - First play-through of each area: expect more shader-compile stutter than on your PC until the caches fill up. Steamdeck users need a fair warning, I was not expecting this game to be this EXTENSIVE when i decided to recompile it, so you may need to play with your settings. Save files from Switch ARE COMPATIBLE. ~/.local/share/MetroidBeyondRecomp/Metroid Prime 4 Beyond_data/save/user/save.bin I had to stop at the second boss cause IT WAS HARD AS FUCK. THIS IS THE FIRST RELEASE. Expect many more updates and fixes as others play the game and report errors. Always check the github for new updates. edit -- added 1.1.0 update support, sol valley ground fixed, added upgrade option

Fix · x-content switch the NVIDIA fallback rung to a model NVIDIA still serves after kimi-k3 was retired github.com/nirholas/three.ws/commit/d989313 · 2026-11-03 · nirholas

nobumin/mtls_manager mTLSの証明書を自動管理するサーバの公開リポジトリです。 Let’s Encryptによってサーバ証明書の管理から解放されたかと思いますが、サーバ間の安全な通信（mTLSを選択）を簡単に管理できるようにしたいという目的で作りました。 docs/playground下に自社サーバ間でmTLS通信する場合の動作確認環境構築設定一式がありますので、どのようなことができるのかを簡単に確認できるかと思います。 （設定などが少し複雑になりますが、他社とのサーバ間通信でmTLSを管理することも可能です。docs/ja/federation-design.mdに記載されています） ”mTLS導入したいけど面倒くさいと”思っていらっしゃる方向けです。 よろしくお願いいたします。

Should be paired to Litecoin as the coin was based on Litecoin. It has nothing to do with XMR. Proof on Github: github.com/muddafudda/Memecoin/blob/master/src/in… t.co

Distributed, Scalable, Auto-Research... Github: iammrduncan/thermite

@thesimaximalist @nvidia Of course! GPUStack supports custom inference backends 💙 Give it a try: gpustack/gpustack

Skills and checker, MIT-licensed: abhishekramkrishna/agent-harness-lab

MOUNT NETWORK DRIVES FROM MAC MENUBAR! 🖧 Keepcoolch SmbmounterPrevents context degradation for Codex by continuously reconciling codebases toward project goals in fresh runs. KeepCoolCH/SMBMounter t.co

What is Git? Git vs GitHub Explained | Version Control System Understand what a Version Control System (VCS) is and learn the key differences between Git and GitHub. In this video, we'll explore why version control is essential for software development, how developers track changes in their code, and how Git and GitHub make collaboration easier. What you'll learn: - What is a Version Control System (VCS)? - Why do developers need version control? - Centralized vs Distributed Version Control Systems - What is Git? - What is GitHub? - Git vs GitHub: Key differences - How GitHub helps developers collaborate Whether you're a college student, aspiring software engineer, or working developer, understanding Git and GitHub is an essential skill. Resources & Links: Notes: RushikeshGhodke/git-and-github-course Complete Git and GitHub Playlist: youtube.com Complete Docker Playlist: youtube.com Connect with me on LinkedIn: linkedin.com If you found this video helpful, subscribe to Build With Rushikesh for more tutorials on DevOps, software engineering, cloud computing, and developer tools. Have questions? Drop them in the comments! [bn] Git কী? Git বনাম GitHub এর ব্যাখ্যা | ভার্সন কন্ট্রোল সিস্টেম ভার্সন কন্ট্রোল সিস্টেম (VCS) কী তা বুঝুন এবং Git ও GitHub-এর মধ্যে মূল পার্থক্যগুলো জানুন। এই ভিডিওতে, আমরা আলোচনা করব সফটওয়্যার ডেভেলপমেন্টের জন্য ভার্সন কন্ট্রোল কেন অপরিহার্য, ডেভেলপাররা কীভাবে তাদের কোডের পরিবর্তনগুলো ট্র্যাক করেন এবং Git ও GitHub কীভাবে কোলাবোরেশন বা পারস্পরিক সহযোগিতাকে সহজ করে তোলে। আপনি যা শিখবেন: - ভার্সন কন্ট্রোল সিস্টেম (VCS) কী? ডেভেলপারদের ভার্সন কন্ট্রোল কেন প্রয়োজন? সেন্ট্রালাইজড বনাম ডিস্ট্রিবিউটেড ভার্সন কন্ট্রোল সিস্টেম - Git কী? GitHub কী? Git বনাম GitHub: মূল পার্থক্যসমূহ - GitHub কীভাবে ডেভেলপারদের কোলাবোরেশনে সাহায্য করে আপনি কলেজের ছাত্র, উচ্চাকাঙ্ক্ষী সফটওয়্যার ইঞ্জিনিয়ার, বা কর্মরত ডেভেলপার, যেই হোন না কেন, Git এবং GitHub বোঝা একটি অপরিহার্য দক্ষতা। রিসোর্স ও লিঙ্ক: নোটস: RushikeshGhodke/git-and-github-course সম্পূর্ণ গিট এবং গিটহাব প্লেলিস্ট: youtube.com সম্পূর্ণ ডকার প্লেলিস্ট: youtube.com লিঙ্কডইনে আমার সাথে যুক্ত হন: linkedin.com এই ভিডিওটি যদি আপনার জন্য সহায়ক হয়ে থাকে, তাহলে ডেভঅপস, সফটওয়্যার ইঞ্জিনিয়ারিং, ক্লাউড কম্পিউটিং এবং ডেভেলপার টুলস-এর উপর আরও টিউটোরিয়ালের জন্য 'বিল্ড উইথ রুশিকেশ' সাবস্ক্রাইব করুন। কোনো প্রশ্ন আছে? কমেন্টে জানান!