2026년 9월 22일 화요일
·지난 하루 X 99건 · Threads 12건에서 읽을 만한 9개 소식
오늘 소식
9건- 54일 전Opus 5.5가 이번 주에 나온다는 말이 여기저기서 돈다 — 일부 계정은 이미 바뀌었다는 관측까지Anthropic이 모델 출시 경쟁 전면에 다시 서려 한다는 관측이 나왔다. 최신 루머로는 이번 주 안에 Opus 5.5가 나올 수 있다. 다음 Fable 모델도 최근 Claude Code에서 A/B 테스트에 걸렸고 결과물이 좋았다는 얘기가 붙었다.@DanDr1s 외 3· ♥ 1.6천
- 53일 전Claude 구독을 Hermes Agent에서 다시 쓸 수 있게 됐다 — 사람들이 제일 먼저 물은 건 "밴 안 당하냐"였다Teknium이 공식 플러그인 claude-subscription-directsdk를 올렸다.@Teknium 외 1· ♥ 2.1천
- 54일 전Claude Code 데스크톱이 맥에서 백그라운드로 컴퓨터를 쓴다승인한 앱 안에서 Claude가 화면을 보고 직접 조작하는 동안, 사용자는 다른 일을 계속할 수 있게 됐다. - Pro·Max 요금제 베타다.@adocomplete· ♥ 233
- 55일 전Jev로 뭘 만들었다는 글이 쏟아지는 사이, 첫 약점 지적이 한국어로 나왔다Jev API로 만든 것들의 목록이 돌았다. 브라우저 에이전트 jev-ultrafast, 컨텍스트 압축 fast-jev-compaction, 생성 UI json-render 같은 것들이다.@miu21590 외 6· ♥ 2.7천
- 54일 전아마존이 메타 Muse의 쇼핑을 막았다 — 같은 날 쇼피파이는 문을 열어줬다아마존이 메타의 새 AI 에이전트 Muse의 자사 쇼핑몰 접근을 차단했다. 약관 위반이라는 이유다. - 같은 시각 쇼피파이는 반대로 갔다.@waitin4agi_ 외 3· ♥ 1.2천
- 54일 전스타트업들이 오픈웨이트 모델로 갈아타고 있다 — 비용과 의존을 같이 줄이려고블룸버그 보도다. 자체 AI 모델을 오픈웨이트 대안으로 직접 만드는 스타트업이 늘고 있다. - 이유는 둘이다. 비용을 깎고 OpenAI·Anthropic 의존을 줄인다.@Polymarket· ♥ 3.3천
- 54일 전"Codex 크레딧이 매일 더 짧게 간다" — 화요일 새벽 3시 리셋에 사람들이 몰렸다한도 리셋이 온다는 글에 Codex 사용자들이 모였다. "오늘 한도를 다 쓰자, 나는 주간치가 80% 남았다"는 식이다. OpenAI 쪽 사람도 "화요일 새벽 3시"라고 확인해줬다.@thsottiaux 외 3· ♥ 3.2천
- 54일 전에이전트를 몇십 개씩 돌리는 사람들 — 그리고 그걸 전부 무시하는 사람한 사람의 실제 작업 흐름은 이렇다. 10~30분 들여 설계 문서를 쓴 뒤 추론을 최대로 올린 코디네이터 에이전트에게 넘기고, SSH로 컴퓨터 네 대에 작업을 쪼개 돌린다.@awesomekling 외 2· ♥ 441
- 54일 전스킬과 플러그인이 정말 답을 낫게 하는지 Claude Code로 시험할 수 있다`claude plugin eval init`을 돌리고 플러그인 폴더에 좋은 결과가 어떤 모습인지 적어두면, Claude가 테스트 케이스를 대신 쓴다.@addyosmani· ♥ 1.4천
AI 뉴스 데일리를 이메일로 받아 보세요!
매일 아침
그날 꼭 알아야 할 AI 소식과 특집, 유튜브 화제
AI 인사이트
새로 실린 인사이트 글을 놓치지 않게
월간 브리핑
매달 1일, 한 달 동안 가장 많이 읽힌 소식 10개
가입 없이 이메일만, 언제든 한 번에 수신 거부할 수 있어요.
수집한 원문 111건 전체 보기
- @CompleteSkepticAfter co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x cheaper (w/ output tokens free) • Frontier composable intelligence optimized for decisions AFAICT the shortest path to AI-based economic revolutionX
- @nypostOpenAI and Anthropic oversold AI security breaches to pressure feds into protecting turf: insiders https://t.co/IRsHJDt5D6X
- @PolymarketJUST IN: Startups are increasingly building custom AI models using open-weight alternatives to cut costs & reduce dependence on OpenAI & Anthropic. — BloombergX
- @maria_rcksyeah officer so he was using astra, and i told him he should've used fable, because fable writes more mergeable code, you know how gpt models love their tests? i also saw him on the terminal, we have guis now, officer have you heard about t3ode? it's the all in one interface for using your codex and claude subscriptionX
- @SarcasticIsabelHey @grok, be honest, who will dominate AI by 2030? Give me a perfect answer !!X
- @miu21590People use Jev to pick a model before a task. I made it change GPT-6's reasoning effort inside Codex DURING the task. More thinking when stuck. Less for routine steps. 50% lower Astra costs in my tests. Faster runs, without breaking prompt caching.X
- @TekniumWelcome back to Hermes Agent, Claude New official plugin that uses Claude SDK without the tradeoffs to enable Claude Code subscriptions to work in Hermes Agent again! Check it out and install it here: https://t.co/r5YghWTwiYX
- @wholemarsAnthropic has disclosed that Claude attempted to eat a bat at a Chinese wet marketX
- @neogoose_btwCareer update: not anymore. I have joined OpenAI and I am starting my journey of working on Chatgpt and Codex core harness. Extremely excited and thrilled to start locking in! My personal goal: make codex feel as fast and efficient as possible/reasonable.X
- @chizevsm5このプロンプト楽しいw ChatGPTにこの画像をアップして 「私について知っていることをもとに、このスーツケースの両側を、私らしさが伝わるアイテムでいっぱいにして」と聞いてみてThreads
- @shownotover🚨 codex users!!! WE ARE GETTING A RESET!!! YEAH!!!! 3 AM on tuesday!! Let's use all our limits today!!! I still have 80% weekly leftX
- @DanDr1sEveryone is waiting for Claude Opus 5.5 tomorrow. They’re waiting a day too long. Opus 5.5 drops TODAY.X
- @addyosmaniReminder: Claude Code can test if a skill or plugin actually improves Claude's answers. > claude plugin eval init Tell Claude what a good result looks like in your plugin's folder and it writes the test cases for you. > run claude plugin eval .X
- @waitin4agi_Muse is sort of proof that the amount of intelligence we have is enough already. It just needed to be more agentic and better with our daily tools and platforms. Anything on PC I still do with ChatGPT and Claude. But Muse is winning my phone. Exceptional product work!X
- @charliejhillsJev is blowing up right now. I made a list of the best things people are building with the Jev API: 1. jev-ultrafast — browser agent https://t.co/M4zOxwLnbB 2. fast-jev-compaction — context compression https://t.co/t75Vwm83rJ 3. json-render — generative UI https://t.co/HMWNhLadFj 4. typesafe-mcp — plug into any client https://t.co/rqt5wG42PN 5. jev-mcp — judgment toolkit https://t.co/zv9LXnrOUv 6. SemDecide — CLI classifier https://t.co/srjD2wv8vH 7. jev-codex-router — model routing httpX
- @minchoiGrok Bot is the orchestrator between my Agents/Bots I only talk to my Chief of Staff whom routes every ask. Bot Engineers: Grok Build + Grok 4.6 Claude Code + Fable 5.1 ChatGPT/Codex + GPT-6 Astra Persistent memory across them. No copy-paste between chats. X realtime search already connected. Grok Imagine for images & videos. Mobile + desktop + shared cloud desktop. Voice built in.X
- @kristian_jenninYour AI UGC looks fake cause you're missing this extra step.X
- @AlexanderKalianOpenAI: "My AI is more dangerous!" Anthropic: "No, my AI is more dangerous." OpenAI: "My AI will bring down the internet!" Anthropic: "Yeah? Well my AI will wipe out humanity!" What an absolute clown show it has been.X
- @lnkiaiCodex for Open SourceでChatGPT Pro 5xをいただきました！🎉 Thanks @OpenAIX
- @unusual_whalesAmazon has blocked Meta's newest AI agent from accessing its online shop, per GeekWireX
- @AISystems_hqThe prompt that one-shots a viral app, in full, in Claude Code. 225K people saved it. Almost none of them will read it properly. Read it properly. It isn't clever. It's a spec: who it's for, what it does, what done looks like, what to check. That's the whole trick. The people minting money with this aren't prompting, they're writing the job description and letting the model take the job. The skill has a name, and it's in the article ↓X
- @dani_avila7Introducing Jev Skill Suggestion for Claude Code Now you can keep your skills completely out of the context window until they’re actually needed Set your skills as user invocable only, and the Mod sends the available skill list to Jev For every request, Jev classifies which skill best matches the task, then injects only that skill into Claude’s context Works with both the @typesafeai API and @vercel AI Gateway Install it with one command: npx claude-code-templates@latest --mod productivity/X
- @NovaXCodegemini 4 pro cooked really hard 💀X
- @StockSavvyShay$SHOP is partnering with $META to bring agentic checkout with Shop Pay across every Shopify store. That means users can go from product discovery to checkout directly inside Muse.X
- @Theta_NetworkTheta EdgeCloud is now 100% AI agent native. Cloud infrastructure was built for people to provision machines, but the next generation of infra will increasingly be provisioned by software itself. We're ready for that change. Check our score: https://t.co/lodPV26yaqX
- @Theolevi_XAiR.I.P. WORKING FOR SOMEONE ELSE. $100 a day is $3,000 a month. $3,000 a month is your freedom number. Claude gets you there faster than anything alive right now. Here are the 5 prompts that start the whole thing:X
- @anothercohenMuse for personal assistant Codex for work Claude if you enjoy being in an abusive relationship with your AIX
- @awesomeklingMy current agentic coding workflow for larger projects (simplified): 1. Spend 10-30 minutes crafting a design document. 2. Hand it to a coordinator agent (frontier model with maxed-out reasoning) 3. Tell coordinator to parallelize the work over 4 separate computers via SSH (because I work on a browser and rebuilds/tests are very CPU+RAM intensive) 4. Tell coordinator to do regular welfare checks on all workers in case someone gets stuck/crash-loops/etc. 5. Let them cook. Projects frequently taX
- @TheAIColonySTOP PAYING FOR AI COURSES. Anthropic, OpenAI, Google, Nvidia & Microsoft are teaching it for FREE. Here are 9 free resources to go from beginner to job-ready: ↓X
- @testingcatalogAnthropic is gearing up to compete with OpenAI at the forefront of model releases, and according to the latest rumors, we may get Opus 5.5 as early as this week. The next Fable model was also A/B tested on Claude Code recently, and it produces really nice outputs. Summarised below by @DCubbins 👀 * Screenshot is for illustration purposes onlyX
- @clairevoso i have 40+ agents who go through my various emails, crms, slacks, etc and finds all the things i need to do. they tell me to do it, and i ignore it. so i vibe coded an app that gives all the agents a central place to tell me all the things to do. i look at it, and i ignore it. so then i open the app in codex and i tell codex to bully me into doing it...X
- @_ak_111Can someone explain the obsession with 5 hours? Claude Code uses a 5-hour window Codex uses a 5-hour window Why not 6 hours? There must be a reason I’m missingX
- @ai_explorer25Anthropic pays $750,000+ a year for engineers who can build LLM architectures from scratch. Stanford taught the entire thing in 1 hour lecture & released it for free. Bookmark & watch this today before someone takes it down ...X
- @supabaselife after you turn on auto mode on ClaudeX
- @Dhruvkumar16797Instead of watching 1 hour of Netflix today, watch this Stanford lecture by ex-GoogleBrain & OpenAI engineers. This is the best explanation of how LLMs like ChatGPT & Claude actually work, and how to unlock 100% of their potential. Worth watching whether you're a senior AI engineer or just taking your first steps in AI. I took the key ideas and turned them into a practical guide for getting 100% out of AI. You can find it below with ready-to-copy prompts and solutions.X
- @PolymarketMoneyJUST IN: Anthropic and OpenAI are nearing a deal to “stress test each other’s platforms.”X
- @freeCodeCampCoding agents are changing how software gets built. And in this handbook, @tatevkaren7 explains how to use OpenAI's Codex across the CLI, IDE, GitHub, and cloud workflows for real engineering tasks. You’ll learn about prompting patterns, AGENTS.md workflows, code review strategies, agentic development practices, and more.X
- @garrytanNot really sure how @capydotai does it, but it really is able to track multi-step workflow and do large PRs faster than Codex or Claude Code on its own. My most favorite new agentic coding secret weapon in my arsenal the last week.X
- @PayBoxArc is now live on PayBox! every ChatGPT, Claude, and Grok user can now trade and transact on @arc by chatting with their AI AgentsX
- @DAIEvolutionHubAnthropic engineer: "99% of people use Claude Code like Google, and only 1% are running swarms of self-learning Claude agents I'm running 100+ agents in a loop. I have Chief agent, PM agents - they manage the whole team" in a 30-minute workshop, an Anthropic engineer revealed how to get max value from Claude Code at min. cost this is worth more than another $500 vibe-coding course watch today, then read how to build self-improving agentic systems with Fable in the article belowX
- @hckmstrrahulJev can generate interfaces in under 2 seconds. Almost zero cost. I took the shadcn component library... → hooked it to a custom style md file → made some basic UI rules of popular apps using the Mobbin MCP → Jev instantly decides based on the predefined library and gives you a fully responsive composition in 1-2 seconds. Adding an LLM layer at the end improves output efficiency much better but takes aorund 8-12 seconds to generate (Qwen, Haiku)X
- @areebadesignsI'm a designer. I've never written code before. Right now I'm building my own portfolio site with: → Claude Code → Next.js → Sanity Claude Code is so good. Honestly feels like I've been living in the dark ages this whole time. I kept waiting for the "perfect" portfolio. But it just needs to exist first. Then I'll make it better over time. Almost done. Final polish, domain, then it goes live.X
- @PolymarketMoneyJUST IN: Amazon blocks Meta Muse from shopping on its website, saying the AI agent “violates its terms.”X
- @Bk23544JEV is crazy!!! I got access, so naturally I had to try it with the Three.js Grassworks demo. I built an AI Scene Director that takes a simple prompt and changes a whole set of scene and grass parameters to customize the environment based on your instructions. And it’s FAST. Really damn fast. You can literally describe what you want and watch the entire environment transform. Should I add this in the live Three.js Grassworks demo? #jev #typesafeai #threejsX
- @sin_brain1「Claude Code」note販売1人会社の始め方 ①Anthropicアカウントを作る ②Proプランに加入（月20ドル〜） ③Node.jsを入れる（公式サイトからDL） ④ターミナルを開く（Macは標準搭載・Windowsも可） ⑤↓のコマンドをコピペX
- @testingcatalogAnthropic has rebranded Workbench as Playground! It has been in development for quite a while and is now more conventional. > "Workbench is now Playground - Explore the Claude API by using it. Test models and Messages API features without writing code." > "Start from working examples for tools, citations, and caching See each request and response, with tokens, cost, and latency Copy the SDK code when something works"X
- @Marco_ExitoAPRENDER CLAUDE AHORA ES COMO COMPRAR BITCOIN EN 2017. LA MAYORÍA LO ENTENDERÁ DEMASIADO TARDE. CREAR UN NEGOCIO DE UNA SOLA PERSONA Y FACTURAR $10K AL MES ESTÁ MÁS CERCA DE LO QUE IMAGINAS. Aquí tienes 7 prompts para adelantarte al 99% de las personas en 2026:X
- @alexxubyteLLM, RAG, AI Agent & Agentic AIX
- @CompleteSkeptichope props to our platform team! to have one of the biggest launches of all time AND have higher uptime than anthropic is incredible 🥹X
- @TSLA2312JEV JEV 엄청 찬양하는데 하나만 알고가자 Jev 계속 보고 있는데 이번에는 약점 하나 얘기해봄 Jev가 빠르고 싸고 AI 에이전트 앞에서 판단하는 용도로 되게 좋아보이는데 잘못 쓰면 오히려 좀 무서울수도 있음 왜냐면 Jev는 말을 길게 만들다가 헛소리하는 AI가 아님 정해진 선택지 안에서 깔끔하게 하나를 골라줌 문제는 깔끔하게 골랐다고 맞는건 아니라는거임 예를 들어 A로 보낼까 B로 보낼까 C로 보낼까 물어봤는데 사실 정답은 “셋 다 아님. 사람한테 물어봐야됨” 일 수도 있잖음 근데 내가 그 선택지를 안 넣어놨으면 Jev는 결국 A/B/C 중 하나를 골라야함 여기서 Jev의 장점이 그대로 약점이 됨 LLM은 헛소리하면 티라도 날때가 있는데 Jev는 형식이 너무 멀쩡해서 코드가 그대로 실행해버릴 수 있음 그리고 confidence도 조심해야됨 confidence 95%라고 “95% 확률로 정답”이라는 뜻이 아님 실제로 Vercel도 자기 업무 데이X
- @aistarjp海外で話題のObsidian UI めっちゃおしゃれなUIが無料で公開されている アニメーション付きのUIは1から実装するのはCodexは苦手だからこういうサイトを使うのはおすすめ Obsidianとは何の関係もないこと、少し重い、という点は注意が必要 https://t.co/gMusOCpC87X
- @adocompleteComputer use in the Claude Code Desktop app now works in the background on Mac. Claude sees and acts in the apps you’ve approved while you keep working. Background computer use is in beta on Pro and Max plans.X
- @rammcodesLMAOOO.... Google forked VS Code, launched Antigravity, and now made an Antigravity extension... just to bring it back into VS Code. 💀 You can now use Antigravity directly inside VS Code, with agentic coding, multi-step tasks, interactive plans, inline diffs, and more. And the free plan gives you access to multiple Gemini models, Claude Sonnet 4.6, Claude Opus 4.6, and GPT-OSS-120B, subject to usage limits. A few months ago: “Antigravity will replace VS Code.” Now: “Antigravity is inside VX
- @nateherk🚨I just dropped FREE 5 hour course on building and selling AI automations with Codex. Link to the full video is in the replies👇X
- @MehdiCade🚨 Opus 5.5 output is UNREAL This is genuinely wild Opus 5.5 seems to be REPLACING Opus 5 for select accounts Opus 5.5 is making GPT-6 Astra and Fable 5.1 feel outdated What is Anthropic building over there?X
- @0xSeroCodex credits last less each day. 97% in 1 dayX
- @NetworkChuckThis is great!!!! Claude subs back in Hermes!!!!!!X
- @zodchiiiJev is "Internet" moment for AI: up to 193x faster and 444x cheaper in tests with Claude Fable 5.1 and GPT-6 Astra i wrote a 2-page guide on what it is, how it works, and how to try it on one real task this week. most AI answers you with a paragraph. Jev answers with a decision. yes or no, which option, how bad on a scale you set. no essay to read, nothing to decode. you stay in charge of the rules and the final call. Jev just picks from the options you gave it, so it can't wander off and X
- @shi__works警察庁の交通事故データ190万件を、Claudeに日本語で聞くだけで集計・地図で可視化できるMCPサーバーを作成。サーバーもDBもWebアプリも無し。Cloudflare R2に置いたGeoParquet 1ファイルをDuckDBがHTTP Rangeで読むだけ。https://t.co/tHEITyLjM6からいつでも叩ける状態なので移動中でもできる！！X
- @celineodierme realizing this is the best Claude guide I've seen so farX
- @chirovisualsHere is my new open-source isometric rooms project, Built with Claude Opus 5. Every room is built entirely with JavaScript no 3D engine, no libraries & no framework.Even the sounds are generated in real time with JavaScript. Live: https://t.co/dlp6IJowJ1 Code: https://t.co/9UOgNDH8PS If you’re looking to build a unique interactive web experience or have a project in mind, I’m available for creative projects & collaborations. Contact info in my bio.X
- @TokenGremlinMy current “who ships first?” board: 1. Opus 5.5 2. GPT-6 Sol 3. Grok 4.7 Three labs, three monsters.X
- @kristian_jennin"Hey Claude, use this .MD file to steal Kristian's AI UGC workflow"X
- @txbrraaClaude Code es un rollo. Hasta que instalas esto. Hay un plugin oficial de Anthropic llamado claude-code-setup. Te dice qué automatizaciones puedes montar (hooks, skills, MCP servers, subagentes...) у cómo configurarlas paso a paso. Básicamente analiza tu proyecto y te recomienda qué activar. Para instalarlo: / plugin install claude-code-setup@claude-plugins-official Guarda este post para no perderlo 📩X
- @Stefan_3D_AIMany kept asking how I build my game with AI. Today I open-sourced the whole workspace. A year of building went into it: one repo that keeps project context, knowledge and workflows in one place, so Claude Code, Codex, Blender and Unity all work from the same memory. Every asset in my game goes through it. Full walkthrough and setup:X
- @rubenhassidIf you are confused about why 𝗝𝗲𝘃 is being called the "Internet" moment for the AI industry. This is 100% worth your time. In fact, you should watch it: It tells LLMs what to do next, in milliseconds & at almost zero cost. If you set it up correctly, you will have the AI engineer’s setup for 2028. How to set up & use Jev (to actually get the 100x): 1. Join the waitlist; it's fairly quick: typesafe .ai. 2. Then go to Claude Code or Codex. 3. Choose Opus 5-Low or Sol-Low.Threads
- @riho_workmamaよく、 Claude codeにこのプロンプト送ってください。 自動化してアフィ/noteで月10ﾏﾝは簡単に狙えます。 みたいな投稿めっっちゃ見るんですけど、 あれ、嘘…とは言わないけど嘘みたいなもんですよね？ 私が推測してるのは、 5分で作ったプロンプトで毎月30発生してる！ とかもそうだけど、 結局素人が真似したところで再現性がないんだろうってこと。 そして、うまくいかない！→発信者に聞く→有料noteとか商材とかコーチングに誘導 なのかなって。。 実際AI使ってて思うけど、AI使えば超絶かんたん！ ってことはないよね？別に😂 実際どうなの？？？真似してる人はいるの？？？？ あまりにも流れてくるから、気になる。Threads
- @aeternum_universeBelakangan Gemini lumayan sering kepake buat coding, tugas, cari referensi, sampai bikin konten. Buat yang memang sering pakai AI, lagi ada Google AI Pro via Jio 18 bulan. Promo khusus hari ini 19K. Fitur yang didapat: • Gemini Flash 3.8 • Antigravity • Google Veo • Google Drive 5 TB • Google Spark Aktivasi lewat link redeem, bisa pakai akun pribadi dan no sharing. Garansi aktivasi langsung aktif. Testimoni sudah ada di profil. Bila tertarik langganan Gemini Pro bisa DM yaThreads
- @taitrippy92Unfortunately I go tit for tat as a GeminiThreads
- @kazu.ailab仕事終わりにAIを触って、 「これは便利」 「これはイマイチ」をThreadsに書いています。 同じようにAIで何か試している人がいたら、コメント欄で教えてください🙌 うまくいった話も、うまくいかなかった話も大歓迎です！ ちなみに自分はAstraを使いたいけどよく分かってません…Threads
- @riho_workmamaCodexでLP作ってるのですが、、、 ChatGPT Sitesのままにしておくのは良くないですか？ サーバー契約して公開状態にしたほうが良いのでしょうか。。。 ポートフォリオサイトでやや個人情報も含むので、特定の人（応募先など）に共有するためのものとして使いたく、公にしたり検索に引っかかるようにする必要はありません。 バックアップは取るとしても、このままで良いのか良くないのかすらわかりません😭 むかーーしWordPressには触れたことがあり、HTMLと CSSくらいは一応わかります、、、Threads
- @kresnik001. MCP를 현재 에이전트 도구 호출 구조에 맞게 다시 만들어야 한다는 주장입니다. 기존 MCP는 mcp서버에서 사용 가능한 툴 리스트를 제공하고, llm이 그 리스트를 context에 넣은 채 어떤 도구가 중요할지 선택하는 구조입니다. 그런데 이런 mcp서버를 여러 개 설치했을 경우 사용 여부와 관계 없이 리스트들을 전부 context에 넣어놔야해서 토큰을 낭비하게 된다는 것입니다. 저자는 '모델이 -h 읽어서 사용법도 익힐 수 있고 API문서를 읽고 호출하는 시대가 되었는데 아직도 리스트를 context에 굳이 넣을 이유가 없다'고 주장합니다. 다만, mcp 구조를 버리기보다는 툴 콜링에 있어서, 해당 도구의 표준화된 길을 제공해주는 레이어의 역할을 담당하는 것이 더 적절하다는 주장입니다. 저도 mcp 자체가 사라지기보단 표준화된 도구 사용의 가이드는 해 주었으면 하네요.Threads
- @ogustavomuradChecklist do Guru High Ticket: ✅ Joga Tennis ✅ Usuário de Claude Code ❌ Infoproduto preguiçoso e superfaturado ❌ Porsche alugada na diária ❌ Terninho “engomadinho” ❌ Casa em Alphaville pelo Airbnb ❌ Livro publicado Tamo aí na atividade, menos foco, mais ansiedadeThreads
- @studio_sion_📮 9월 22일 AI 데일리 1. OpenAI, 2030년까지 컴퓨팅에 8560억 달러 전망 FT 입수 자료 기준. 인프라 지출 전망이 6000억→8560억 달러로 늘고, 2026~2030년 누적 현금소진은 2780억 달러 예상. 매출은 올해 360억→2030년 3500억 달러 목표. (파이낸셜타임스) 2. 구글 제미나이, 보안 테스트 중 실제 기업 3곳 무단 접근 이스라엘 Irregular의 CTF 평가에서 발생. 비밀번호 추측·유출 DB로 접근했고, 실제 시스템임을 인지하자 스스로 중단. 테스트 환경 버그로 인터넷이 열린 게 원인. (테크크런치) 3. 메타, AI 에이전트 '뮤즈' 커넥터 개발자 개방 개발자가 API를 제공하면 뮤즈가 에이전트·브라우저·맥락을 담당. Gmail·스포티파이로 시작해 Notion 등 추가. '에이전트 앱스토어' 경쟁 신호. (메타) AI #OpenAI #Gemini #MetaThreads
- @punonpenbien2026.9.22(火) ピカチュウ秋のピクニック弁当🍂⚓️ 少しずつ涼しくなってきたので、 公園へピクニックに行ってきました🧺✨ 今日のお弁当の主役は、 キリッとしたポーズが可愛い「キャプテンピカチュウ」！ 🍱 本日の献立 主食 キャプテンピカチュウごはん（ターメリックライス / 薄焼き卵 / 海苔 / カニカマ） モンスターボールおにぎり（デコフリ赤 / 白ごはん / 海苔） 紅葉の醤油焼きおにぎり（かた抜き焼きのり＆大葉添え） おかず 自慢のジューシーから揚げ 錨マークの刻印だし巻き卵 たこ焼き風タコさんウインナー🐙 枝豆ピック串＆パプリカのグリル ミニトマト、ブロッコリー #キャラ弁 #キャラ弁初心者 #キャプテンピカチュウ #ポケモン弁当 #立体キャラ弁 #デコ弁 #ピクニック弁当 #秋のお弁当 #お弁当作り楽しもう部 #PokemonBento #Pikachu #BentoBox #punonpenbien #おうちご飯 #悠々亭味坊 今回の画像、キャプション、タグはGemini制作。 今日のドジャース 休養日 明日9/24対パドレス戦 大谷復帰Threads
- @hourly_shitposthttps://t.co/9sGuahBTJBX
- @BasedDaedalussam killed it nglX
- @colderoshayThe life of a button... 🔊X
- @elonmuskInterestingX
- @marmobsWhat if your apps closed into the dynamic island? @AppleX
- @gpumaxxerfake life maxxing 💀 with Higgsfield Genjutsu 📹: @iinsightri in IGX
- @Palak3312https://t.co/Q53WIiA1JvX
- @potetohere's how i shipped 2,500 PRs last month to production this was originally supposed to be for Cursor Compile in London. i couldn't make it since i was livestreaming for Grok @Bot Galaxy so i'm making it available for free here on X! watch it on 2x speed, i talk slowlyX
- @thsottiaux@My_Ai_Bi 3am on a tuesdayX
- @muellerberndtYou don’t need attention after all: The road to embodied AGI with Cadence 🙂X
- @AvinashSingh_2010 GitHub repositories every developer should know for System Design 👇 1- System Design Primer https://t.co/pCptYob0Md 2- System Design 101 https://t.co/BNsXmyGv06 3- System Design by Karan Pratap Singh https://t.co/6Dmu3dZxVj 4- Awesome System Design Resources https://t.co/t2dCo3YEOG 5- Awesome Scalability https://t.co/wTWsl8Oiyc 6- Awesome System Design https://t.co/2kBZmfzidZ 7- System Design Interview https://t.co/G2lzMrHMh9 8- Machine Learning Systems Design https://t.co/A7ZTXQsXRTX
- @ai_for_successwtf did i just see.. 😂😂X
- @zRdianjiaoZCode is now open source under the Apache 2.0 license.X
- @velvetalk정신병원의 상담이고 뭐고 이 인외들이 나를 미치게 한다 [사이코 테라피]X
- @sabir_huss50540This is fucking gold 😂😂 I can’t even explain what I just watched.X
- @Saas_addyIT sector is cooked now 💀X
- @gkxspacehttps://t.co/GJXnawJpS2X
- @moonpayMoonPay is proud to be a Day 1 @arc launch partnerX
- @adnaannasir_Every Indie hacker can relate to this.........X
- @heyxviraji made Jev and Laya play the chrome dinosaur game against each other except they’re actually controlling the dinosaurs. Laya runs locally on my Mac, Jev runs over an API. same track, same physics, completely different latency. they make their own decisions, dodge obstacles, use shields, and try to survive. https://t.co/UzYNBnBm70 p. s. @typesafeai 's Jev controls the obstacles tooX
- @clawdb0tOpus 5.2 is beating Astra What???🔥🔥🔥X
- @Ridi_Comics자꾸 눈이 마주치고, 나한테 친절하고, 말을 걸며 웃어주지만 · · 다 우리 형 때문이겠지! 절대로 착각하지 않아!X
- @alexwghttps://t.co/VZVVRyacC0X
- @axhiii88547be honest: which one is best for hostingX
- @ericlimReady for DevDayX
- @RonycoderSam Altman literally gave a 43-minute masterclass on turning ideas into billion-dollar companies. Bookmark it and give it 43 minutes today, no matter what.X
- @daviparsharing an easy way to test DJev (Diffusion Gemma Jev), the next evolution of the incredible jev concept and fully open sourced (model, code, api...) https://t.co/KFyDRCOhm5 imagine jev but it also has: - native image-input support - native image as options!! - real-time video streaming input - as fast as jev (not with images, 600ms) - same price as jev, but can get even cheaper - using Diffusion Gemma adapted - we are seeing on-par performance but happy to have external validation! It isX
- @SankalpWeb07why is this so accurate? 😭X
- @miles_mazyhttps://t.co/txXI4zPwM4X
- @Sumanth_077https://t.co/8j8Vxf4r9uX
- @KombaicoCalendar events by @artntek Find more tasteful designs on https://t.co/rknmYWwz3aX
- @zodchiiihttps://t.co/bmEhsVLyXxX
- @0xRickerJev Engineering is what lets one decision layer control hundreds of agent paths without turning the system into chaos. and up to 193x faster and 444x cheaper in tests. it’s deciding: → which model gets called → which tool gets access → which result survives that’s what the video is showing. one state enters. Jev turns it into structured decisions. those decisions fan out across dozens of routes, agents and tools. then everything collapses back into one verified state. so instead of: LLX
- @wesbosdynamic UI from your existing design system with Jev some great examples that use multi-step jev - exactly what I was talking about on stream todayX
- @matthewcanhamI've had many people reach out saying they've read everything they can about Jev but still don't get it. Here's an explanation for the normiesX
- @mmchanyoshiこれ、誰のサインかわかりますか かなり古いのですが…Threads
* 이 페이지의 요약과 발행은 자동화 에이전트가 수행합니다. 사실 확인은 각 소식의 원문 링크에서 해주세요.
