AI Coding · Launch videos on X
Product videos, demos and walkthroughs shared on X in this niche, newest first, each linking back to the original post with its live view count.
89 launch videos in AI Coding over the last 7 days.
Launch videos
65.2K14:46
412.9K0:25holy sh*t. Official Anthropic tip for Claude Code: stop burning Opus 5.5 context on tasks Haiku 5.5 can swarm, while Sonnet 5.5 builds put the architect on call with /advisor claude --advisor opus --subagents haiku Sonnet 5.5 drives the primary session on high effort, writing diffs and running test suites Haiku 5.5 subagents swarm the repo in parallel at 340 tok/s, handling file discovery and spec docs lookup Opus 5.5 stays on call in the background as the advisor, only stepping in at three critical moments: → before a plan locks: does the implementation plan miss auth invariants or schema contracts? → when a test breaks twice: are we fixing the root cause or falling into a recursive rabbit hole? → before calling done: did the full diff introduce hidden regressions or break pre-flight rules? Opus 5.5 reviews. Sonnet 5.5 builds. Haiku 5.5 swarms Jev engineering runs the identical pattern one layer down: mechanical decisions that need no reasoning (which file to open, which tool to
—2:34📣 Coming soon to GitHub Copilot: intelligent local model routing. We are excited to announce the next step in our Project HydraFusion vision. GitHub Copilot will soon be able to automatically route tasks to a local model when it's most suitable, helping you save on AI credits. https://t.co/lvjtnBswex
765.8K0:30Haiku 5.5 is now available in the Claude Platform and Claude Code. On average, it costs around 75% less to run than Haiku 4.5. It pairs well with Opus 5.5 or Sonnet 5.5 as a subagent. Use it for high-volume, cost-sensitive tasks like summaries, compactions, or database queries. https://t.co/RB9xCWo4Db
68.9K2:27I built an operating system where the AI agent isn’t another app. It’s part of the OS. Today, Herald OS is open source and you can install it yourself 🚀 It’s built around Hermes Agent by Nous Research. Talk or type to Hermes and it can work across your entire computer: • Open and use apps • Find, organize and work with your files • Remember what matters • Run routines and automations • Build software while you watch • Understand what’s happening on your screen • Work across the machine instead of being trapped inside a chat window And when something is about to make a change, you approve it first. 🎙️ Talk to your computer Say “Hey Hermes” and just talk. Native OS commands don’t need to burn model tokens. When you actually ask the agent to think, build or perform a task, that’s when Hermes gets involved. ⚡ Everyday superpowers • Select any part of your screen and ask Hermes about it • Pick a colour from anywhere • Read a QR code from your screen • Copy text from anywhere on
336K0:40
67.4K1:00
54.3K1:04here's how to build a game in Unreal Engine with Claude Opus 5.5... the trick is letting Opus work inside the live editor, but there's a smart way to set it up: - step 1: Unreal Engine 5 + the Third Person template, so you can walk and jump from minute one - step 2: enable Unreal MCP and AllToolsets in the editor, start the server - step 3: install Epic's plugin in Claude Code and have Opus read the project before it changes anything - step 4: one prompt builds a playable base level: a valley, a dirt path, a town at the end, 3 glowing collectibles - step 5: test the whole route with the default mannequin before any custom art - step 6: bring in a rigged character on the UE5 Mannequin skeleton, Opus swaps it in and wires walk, run and jump - step 7: import your forest and town assets, Opus places them along the route - step 8: refine in small batches: vegetation, lighting, a minimap, the town entrance and the rule that saves the project... a successful tool call doesn't mean the lev
98.8K0:12
147.2K0:15Introducing BlitzClean ! A free, open-source RAM monitor + disk cleaner for Mac I built it based on 3 things: - I kept running out of storage - I never knew what was eating my RAM - Memory Clean 2 stopped working for me on the macOS 27 beta (Golden Gate) What it does: - CPU, RAM and free disk in the menu bar - Ranks apps + AI agents (Claude Code, Codex, Cursor) by RAM, so you can pause or quit the ones you forgot about - Finds what fills your disk - Cleans caches, node_modules, simulators, Docker and old worktrees (you review everything first) MIT, runs 100% locally https://t.co/SLkJ2z5Ut5 Made the video with Opus 5.5 medium, what do you think ?
491.6K18:00El ingeniero que creó Claude Code acaba de publicar un vídeo de 28 minutos sobre cómo escribir prompts que funcionan de verdad. He visto cursos de 300$ que no cubren ni lo que él enseña en los primeros 10 minutos. ( El resto del vídeo está en los comentarios. ) Archivos CLAUDE.md, atajos de memoria, sesiones en paralelo, técnicas de prompting… Todo en un solo vídeo, y totalmente gratis. Da igual si eres developer, si estás empezando o si llevas meses usando Claude: te puede servir muchísimo. De esos hilos que te arrepentirás de no haber guardado. 🔖
67.2K0:24I'm sharing with you one of the best papers I've read on Harness Engineering it takes apart Claude Code (Opus 5.5), Codex, Gemini CLI and more, and boils all of them down to one line: agent = model + harness what you walk away with: > what a harness actually is > how today's coding agents are built under the hood > the patterns every one of them repeats > where the whole category is heading > their checklist for building your own the model gets the headlines. the harness decides whether the agent works loop and graph engineering on top of it - in my article below
428.8K0:40I've been working on a skill that makes better HTML plans in Claude Code. It uses simple language, shows code snippets, surfaces questions & makes mockups. Linting reduces the normal failure cases that Claude runs into. Would love your feedback before shipping more broadly! https://t.co/U2Gb1K37vl
36.1K0:24this is the list nobody wants you to bookmark. 10 GitHub projects where AI agents actually make you money not another framework zoo. client work, content, leads, voice, trading. -> n8n · AI workflow automation you sell to clients on retainer. the agency backbone -> https://t.co/BBZfRhkSaO -> Dify · ship chatbots, RAG apps and agent workflows for businesses without a dev team -> https://t.co/yzg8eMkSol -> Firecrawl · any website -> clean structured data. scraping gigs, lead lists, datasets -> https://t.co/vnKNOKPbS2 -> browser-use · agents that click through the browser for you. sell the automation, not the hours -> https://t.co/aff20DFcJu -> TradingAgents · multi-agent trading firm in a repo: analysts, researchers, risk desk (research only - not an atm) -> https://t.co/wGJ1ROMRvm -> screenshot-to-code · screenshot -> clean HTML / Tailwind / React. landing pages in minutes -> https://t.co/neD3z2puLK -> Marketing Skills · CRO, copywriting, SEO and growth skills for Claude Code.
5.2M18:00L’ingénieur qui a construit Claude Code vient de publier une vidéo de 28 minutes sur la façon d’écrire des prompts qui fonctionnent vraiment. J’ai vu des formations à 300 $ qui n’abordent même pas ce qu’il montre dans les 10 premières minutes. ( La suite de la vidéo est dans les commentaires. ) Fichiers CLAUDE.md, raccourcis mémoire, sessions parallèles, techniques de prompting… Tout dans une seule vidéo, et entièrement gratuitement. Que tu sois développeur, débutant ou que tu utilises Claude depuis des mois, ça peut vraiment t’être utile. Le genre de thread que tu vas regretter de ne pas avoir mis en signet. 🔖
49.9K0:24this might be the most useful AI agent paper of 2026, and it's certified f*cking gold Wavestone AI Lab took apart Claude Code, Codex and 9 more agents every good one runs on the same 7-part harness agent = model + harness all 11 use the same 7 parts. only the size changes the 7 parts: > loop - from a plain while loop to fully replayable runs > LLM layer - one prompt template up to 29 provider profiles > tools - bash only up to 43 typed tools on demand > memory - full history up to notes kept across sessions > safety - a step limit up to rules + reviewer model + sandbox > orchestration - Aider skips sub-agents, others fork them > extensions - SKILL.md in 9 of 11, MCP in 8 it ends with a 90-line harness covering all 7 the cleanest starting point for your own agent the pages worth your time: > p.4 - the 7-part map > p.12 - 3 loop types, Claude Code and Codex as examples > p.53 - why nobody uses frameworks or embeddings > p.67 - 18 design rules > p.71 - the 90-line harness to copy
66.9K0:32I don't understand why everyone isn't doing this yet. Anthropic's own Claude Code docs show how to run a whole team of Claudes, while Opus 5.5 only touches the plan and the merge the whole idea: agent teams in Claude Code one lead, separate teammates in their own context windows, one shared task list, and they message each other directly the lead: Opus 5.5 on high, splits the work, writes the tasks, merges at the end the builders: Sonnet 5.5, one owns client/, one owns api/, never the same file the adversary: Fable 5.1, never writes code, only shows up at three points: → before an interface locks: do both sides agree on the contract? → when a test fails twice: is it fixed or just hidden? → before a task is marked done: what breaks it? Sonnet 5.5 builds. Fable 5.1 attacks. Opus 5.5 merges Jev engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to Jev in under half a second, and the team only argues about the ones
97.8K0:36This dinosaur doesn't exist 35 seconds. 5 moves. 8.3M views... Toddler cartoons like this are the quiet monster of YouTube – a breakdown of the format did 1.86M views here Here's exactly how this one was made: 1. The song. One prompt to Lyria 3 Pro with my own lyrics - 35 seconds, vocals and ukulele. The first take skipped "Three!" in a counting song. Whisper caught it, take 3 sang every line 2. The mascot. 4 concepts, picked Dino. Nano Banana Pro in Higgsfield put him in one playroom for every line: hello, jump, clap, spin, touch your toes, bye-bye 3. The singing. I cut the song piece by piece and gave each piece to Seedance 2.5 as an audio reference. He opens his mouth on every word and dances in the gaps 4. The edit is code. Claude Code wrote it in HyperFrames: every word lights up when it's sung, a star hops from word to word, giant numbers pop and fly into a 5-slot counting board. The outro counts the board back: one, two, three, four, five Then every scene is an ad slot:
70.3K0:40yesterday i showed claude code using codex's computer use, today i got some more fucking fire for you claude can also drive codex's official chrome extension, and it controls chromium browsers beautifully all in the background, without the "allow" popup chrome devtools mcp throws at you every time a new agent connects it even handles more than one browser: each install of the extension gets its own instance id, so claude knows which is which i have helium as my main and chrome for work, and claude code now uses both first, install the chrome plugin in the codex app and add openai's chatgpt extension to your chromium browser (or all of them, if you have a few) then give this to your claude: "Set up Codex's Chrome extension for you on my Mac. I've installed the Chrome plugin in the Codex app and the extension. Use Codex's computer use MCP server as user MCP "codex-cu": keep it if it's already registered, otherwise copy the "cua_repl" entry from ~/.codex/plugins/cache/openai-bundle
55K12:32🚨 أنت مش محتاج تبني شركة ضخمة عشان تعمل عشرات الآلاف من الدولارات شهريًا. واحد بيستخدم Claude Code + YouTube وبيحقق حوالي $98,000 شهريًا. 🤯 والأغرب ؟هو مش بس بيعرض النتيجة… شرح الـworkflow كامل خطوة بخطوة في أقل من 15 دقيقة. 👀 لو بتستخدم Claude Code أو مهتم بالـAI، الـ15 دقيقة دول ممكن يغيروا طريقة تفكيرك في صناعة المحتوى والربح منه. 🔖 Bookmark الفيديو وشوفه قبل ما يضيع منك. https://t.co/vpI3VskRBB
37.3K0:11Claude Code es un lío. Hasta que instalas esto. Hay un plugin oficial de Anthropic llamado claude-code-setup. Te dice qué automatizaciones puedes montar (hooks, skills, MCP servers, subagentes…) y cómo configurarlas paso a paso. Básicamente analiza tu proyecto y te recomienda qué activar. Para instalarlo: /plugin install claude-code-setup@claude-plugins-official Guarda este post para no perderlo 🔖
52.2K0:35METÍ 300$. LOS CONVERTÍ EN 14.000$. Y NO PROGRAMÉ NI UNA LÍNEA. cogí un repo de GitHub open source, le dediqué dos horas en Claude Code para limpiarlo, y le di a ejecutar pensando que iba a fundir la cuenta en una hora. se llama 5min-btc-Polymarket. opera mercados btc de 5 minutos en Polymarket entra cuando quedan ~2 minutos para el cierre confirma que btc ya se ha movido 70-100$ en el intervalo opera siempre a favor del momentum, nunca en contra 300$ → 14.000$. sin tocar el código. → nunca predice el futuro, lee uno que ya pasó → compra esa posición entre 0,80$ y 0,99$ → se liquida a 1$ minutos después → cubre solo si el mercado se desequilibra a 95/5 como comprar un boleto de lotería después de que ya hayan cantado los números. os dejo el repo abajo.
676.4K1:16CLAUDE CODE AHORA PUEDE CLONAR LA INTERFAZ DE CASI CUALQUIER SITIO WEB EN MINUTOS existe una plantilla open source que analiza cualquier web y la convierte en un codebase limpio de next.js Le das la URL y tu agente analiza: → interfaz y componentes → colores y tipografía → espaciado → activos como imágenes e iconos Después reconstruye cada sección por su cuenta, con agentes trabajando en paralelo. Funciona con Claude Code, Codex, Cursor, Gemini, Copilot y más de 10 agentes de programación. Más de 35k estrellas en GitHub, licencia MIT y gratis. Resultado: lo que antes eran días de maquetar una interfaz ahora son minutos. Esto cambia el juego para quien quiere migrar una web vieja, recuperar código perdido o aprender cómo está hecha una gran interfaz. Repo en comentarios 👇
43.6K0:31这次真被 Opus 5.5 震到了，瘫坐在椅子上那种 🤯 最近在 vibe coding 一个 proxy 管理面板，做了几套主题，缺个点阵字体，就顺手让 Opus 5.5 设计一个。 结果它很快交出来一款可变点阵字体，还给自己起了名字叫 PUNCTUM。字重 100–900，圆度 0–100，一共 108 个字形，把圆度从 100 拉到 0，圆点会一点点变成方像素。 它还顺带做了一个很专业的字体样例页，把这个字体放进各种老设备的场景里：电梯的 LED 楼层灯、卡座的 VFD 屏、热敏小票、掌机 LCD、火车站发车牌。 我又跟它来回调了几轮，现在的效果放以前，我觉得得是很厉害的字体工作室才做得出来。视频是它做的字体发布样片。 字体还在继续完善，之后会连同整个制作过程一起开源到 GitHub，有兴趣的可以先关注一下。
2M27:52el ingeniero que construyó Claude Code acaba de publicar un video de 28 minutos sobre cómo escribir prompts que realmente funcionan he visto cursos de 300$ que no cubren lo que él muestra en los primeros 10 minutos archivos CLAUDE.md, atajos de memoria, sesiones paralelas, patrones de prompting todo en un video y completamente gratis funciona seas desarrollador, principiante o alguien que lleva meses usando Claude
25.7K0:22CLAUDE CODE CAN NOW TURN HIGGSFIELD INTO A ONE-PERSON AI VIDEO STUDIO. install this 19 skills, connect Claude Code to Higgsfield, then tell Claude what you want to create. run /ugc-hot-girl and let it create the character, generate the visuals, and produce the Seedance video for you. Claude controls Higgsfield through Playwright. you just review the output and make changes when needed. this is useful if you want to make UGC ads for brands without spending hours doing everything manually. you could even turn the workflow into a service and sell the finished content. if you’re building toward the solo creator income path, save this. repo below 👇
42.2K0:23
41.2K2:39
72.4K9:58Anthropic engineer: "90% of our engineers were already running self-improving loops Now everyone is moving toward agentic graphs" "Prompting is basically finished" In just 10 minutes, she builds her complete Claude Code setup live from an empty terminal Agents → Loops → Graphs → Self-Improving Systems Prompting was the old workflow Graph engineering is the next one This 10-minute breakdown is worth more than most $1,000 agent engineering courses Watch it today Then save the full guide below before everyone catches on ↓
224.2K0:15Absolutely recommend using Cache Control in Claude Code Probably one of the Mods that will save you the most tokens and help keep your sessions going much longer It adds this bar for Claude’s 5 minute cache and sends notifications to help keep the cache warm Install command: npx claude-code-templates@latest --mod observability/prompt-cache-control
321.9K0:24I don't understand why everyone isn't doing this yet. Anthropic released a guide for Claude that makes work much more efficient while burning a fraction of the tokens Main Idea: Agent Teams in Claude Code Instead of burning the expensive Opus 5.5 context on every line of code, tasks are delegated across an autonomous swarm: Architect (Opus 5.5, high effort): Plans the structure and handles final PR merges Developers (Sonnet 5.5, medium effort): Write code inside isolated git worktree environments (UI/UX, backend) Adversary / Critic (Fable 5.1): Steps in at contract boundaries, recurring test failures, and pre-PR audits to catch vulnerabilities Script Layer (Jev): Handles mechanical tasks (opening files, CLI invocations) in 16ms without involving LLMs Launch the team via terminal: claude --teammates ux,backend,adversary Configure my Claude Code workspace for autonomous agent teams: In ~/.claude/teams, generate missing roles: set workers to Sonnet (medium effort) in se
341.3K0:40Your AI isn't bad at design. It has just never seen good design. I've tested 100+ vibe coding resources this year. These are the 40 I actually kept, grouped by what you're stuck on. Bookmark this 🔖 1. 👑 https://t.co/RlChWqggj8 Why it's different from everything below: - Every one is a complete prompt, not a snippet you wire up yourself - 3 steps: copy the prompt, paste it into your AI, launch it live - 300+ award-winning prompts, all in one place - Full source code to download and edit - Works with Claude Code, Cursor or v0 Built for founders and freelancers who want award-level sites without a design team or a week of tweaking. If you only open one link in this post, make it this one. 2. https://t.co/zii2nKAKiq - design systems as markdown your agent can read 3. https://t.co/mHD5eSDPKY - ready-made prompts for dashboards and landing pages 4. https://t.co/iRLY9tbhpW - curated high-end sites 5. https://t.co/hOf1AxeaT7 - real UI inspiration mapped to prompts 6. https://t.co/bhB2Gf
16.7K0:28🚨 خبر مفرح للمهندسين: توليد المجسمات والتصاميم ثلاثية الأبعاد (3D/CAD) بالنص فقط أصبح متاحاً رسمياً ومحلياً! 📐💥 🔴 تم إطلاق إضافة text-to-cad بشكل رسمي لمنصة Codex، لتتحول الأفكار البرمجية والهندسية إلى مجسمات ثلاثية الأبعاد قابلة للتصنيع بضغطة زر! • أهم المميزات والقدرات: دعم استخراج الصيغ الهندسيّة القياسية مثل (STEP, STL, 3MF, GLB). تحسين التصاميم لتناسب الطباعة ثلاثية الأبعاد (3D Printing)، الشيت ميتال (Sheet Metal)، مخرجات الـ CNC، والقوالب (Injection Moulding). ربط فوري وسلس مع خدمات وتطبيقات مثل Bambu و SendCutSend لتصنيع مجسمك مباشرة! 💡 الميزة الأهم: مجانية ومفتوحة المصدر 100%، وتعمل محلياً بالكامل داخل تطبيق Codex Desktop بدون الاعتماد على سيرفرات سحابية خارجية! • طريقة التثبيت السريعة بـ Codex CLI: codex plugin marketplace add earthtojake/text-to-cad codex plugin add text-to-cad@earthtojake (أغلق تطبيق Codex وأعد فتحه لتكتمل الإضافة) ضع التغريدة في المحفوظات لتجربتها في مشروعك الهندسي أو الطباعي القادم 📌👇 https://t.co/4PUwQmYrVt
57.7K0:06リサーチ時間90％短縮しますw Claude Codeが、𝕏・YouTube・Reddit・公式の入口がないサイトまで、ほぼどこからでもデータを持ってこれるようになりました。 GitHubに出たばかりの、誰でも無料で使えるツール3つがその超能力をAIに与えます。 1. Agent-Reach 𝕏・YouTube・Reddit・GitHubを1か所にまとめて読める 2. Patchright Enhanced Playwrightで公式の入口がないサイトの通信を拾って、スクリプトでデータだけ抜き出す 3. Scrapling どんなページにも使える万能の収集ツール やることは、欲しいデータを伝えるだけ。 ↓ AIがコードを書く ↓ 勝手に集めてくる ↓ 結果が返ってくる 実際に頼める例はこれ。 「𝕏の100アカウントの先月分の投稿を集めて、伸びてるテーマを教えて」 「この商品のRedditとYouTubeのレビューを集めて、不満をまとめて」 調べ物を任せられる自分の分身が、 寝てる間も手放しで動いてくれる状態です。 これまで人が何時間もかけて見て回ってたものが、頼むだけで終わる時代になりました。 なんちゃってリサーチで時間が溶けてる人、いますよね？w 各ツールのURLはリプに貼ってます。
62.2K27:52Claude Code 工程师刚刚放出了一个 28 分钟的实操视频 毫无保留的教你如何真正榨取 Claude 300% 的生产力： 从基础的如何配置完美的 CLAUDE.md、到持久化上下文记忆、多线程并行会话，以及那些没人告诉你的高阶 Prompt 逻辑。 我花大价钱买过的那些课，干货密度可能连他头 10 分钟的一半都达不到。 而且最爽的是：全免费。 不管你是刚学编程，还是像我一样整天在终端里切 Tmux 疯狂写全栈项目… 看完这个，你会用一种全新的上帝视角去管理你的 AI 编码工作流。 强烈建议码住。 以后你每次重构项目架构的时候，都会回来翻看无数遍。 https://t.co/7kR8rzCAmL
58.5K0:25【事件】 Claude Codeが作業している間だけ、同じくClaude待ちの人たちとDoomで対戦できるプラグインが海外で大バズ😳 https://t.co/hiEEi4lKT3 名前はーーintermission。作ったのはJarrod Wattsで、GitHubで無料公開中 これ、何がヤバいかというと👇 ・Claudeが2秒以上動き出すと、横にDoomの画面が開く ・対戦相手は全員、自分のClaudeの作業待ちをしている人 ・Claudeが終わると3秒カウントして作業画面に戻される ・許可の確認が来たらすぐ戻され、答えるとまた試合へ つまり ❌ Claudeの作業待ちはスマホを眺めて終わっていた ⭕ intermissionで待ち時間がそのまま対戦の時間になる 「Claude Codeが動いてる間、みんな何してる？」 → 同じく待ってる人とDoomで撃ち合ってた
43K0:32Official Anthropic tip for Claude Code: stop burning Opus 5.5 on work Sonnet 5.5 can do, while Fable 5.1 sits idle hand the grunt work to Sonnet 5.5 subagents and put Fable 5.1 on call run /advisor fable Opus 5.5 plans and merges Sonnet 5.5 subagents read, edit and run the tests Fable 5.1 reads the full session and only speaks up at three points: → before a plan: is this the right approach? → when the same error comes back: am I digging in the wrong place? → before "done": what did I miss? Sonnet 5.5 builds. Fable 5.1 reviews. Opus 5.5 ships Jev engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to Jev in under half a second, and the big models only see the ones that split - the full tree > Opus 5.5 on high runs the main session > explorer reads the code on Sonnet 5.5 > worker edits and runs tests on Sonnet 5.5 > researcher pulls the docs on Sonnet 5.5 > all three on medium > Fable 5.1 on call for main and every
42.9K0:27Un tipo japonés descubrió el verdadero apalancamiento de Claude Code antes que nadie. Él simplemente instaló “Find Skills”, describe lo que quiere lograr y el sistema le propone las habilidades perfectas. ¿El resultado? Su sistema automatizado de YouTube que está teniendo éxito gracias a eso. https://t.co/0yGXNSU5qi
61K0:15
245.6K0:20This is f*cking insane. This tip saved me thousands of dollars. run Opus 5.5, Sonnet 5.5, and Fable 5.1 together, and stop burning Opus on work it was never needed for. the whole idea in one line: the strong model plans, the mid-tier model executes, Fable stays quiet until it's actually needed. roles, broken down: Opus 5.5, high effort, owns the plan and ships the final code Sonnet 5.5, medium effort, splits into explorer (reads the codebase), worker (edits files, runs tests), researcher (pulls docs) Fable 5.1, called through /advisor fable, reads everything happening in the session but stays silent unless something's actually wrong three moments where Fable speaks: → a plan goes out: is this actually the right call? → the same failure shows up again: is the search going nowhere? → the task gets marked finished: did something get skipped? Jev engineering does the same thing one level down. the forks that don't need real thought, which file, which tool, keep going or stop, go
104.9K2:25
28.9K0:15📝 Google Maps APIに頼りがちだった「場所検索」に、かなり面白い選択肢が出てきた GISエンジニアの大橋直樹さんが、日本全国の施設を検索できる「OpenPOI API」を無料公開 これがなかなかすごい！ ・日本全国 約337万件 ・APIキー不要、登録不要 ・無料、商用利用OK ・検索結果を自分のDBに保存できる ・「すたーば」→「スターバックス」のような表記ゆれにも対応 ・MCP対応でClaudeなどのAIから直接検索できる POIというのは、ざっくり言えば「店や施設などの場所データ」 Overture Maps 約249万件と、食品営業許可・届出データ 約88万件を統合していて、例えばClaude CodeにMCPでつなげれば、 「東京駅から300m以内のラーメン屋を探して」 みたいな依頼から、施設名、住所、緯度経度までAIが直接取ってこれる これ、非エンジニア目線でもかなり可能性がある 営業リストのたたき台を作る、地域のお店マップを作る、自社サイトに店舗検索を付ける、AIエージェントに「近くの店を探す」能力を持たせる こういうものを作る時に「場所データをどうする？」という面倒な部分を一気にショートカットできる もちろんGoogle Mapsの完全代替ではなく、口コミ、写真、営業時間、ルート案内などはなく、カテゴリ不明の施設もまだ多い でも「オープンデータ × MCP × AI」という組み合わせはかなり面白い こういうAIが自由に扱える“公共データの道具化”は、これから一気に増えていきそう https://t.co/wETI6EYqYX
46.7K0:13Your CLAUDE. md is making Opus 5.5 DUMBER. Stuff like “double-check your work” and “think step by step” was advice for yesterday’s models. On Opus 5.5, it’s drag slower responses, weaker output. Lance Martin over at Anthropic just pushed the remedy. Open Claude Code and run: /claude-api prompt-audit It sweeps your skills, CLAUDE. md, and prompts. Strips the deadweight phrases that kneecap the model. Then rewrites what’s left using the official Opus 5.5 guidance. Takes five minutes. Do it tonight. Save this now, or you’ll swear you were going to later.
47.2K0:40I've tested 100+ vibe coding resources this year. These are the 25 I actually kept, grouped by what you're stuck on. Bookmark this 🔖 1. https://t.co/RlChWqggj8 (mine) - 200+ complete build prompts for 3D, scroll-driven and cinematic websites. - Each one is a full design brief, not a component you drop in. - Copy the prompt, paste it into Claude Code or Cursor, and you get a whole animated site in minutes, not a blank page and a lot of guessing. - Built for founders and freelancers who want award-level sites without a design team 2. https://t.co/zii2nKAKiq - design systems as markdown your agent can read 3. https://t.co/mHD5eSDPKY - ready-made prompts for dashboards and landing pages 4. https://t.co/iRLY9tbhpW - curated high-end sites 5. https://t.co/hOf1AxeaT7 - real UI inspiration mapped to prompts 6. https://t.co/bhB2GfGIix - 2,000+ real product styles with typography 7. https://t.co/ZNFmqbqGsp - 2,600+ examples of how top design systems solve the same element 8. https://t.co
70.6K27:57BREAKING NEWS! The engineer who built Claude Code from scratch just released a 28-minute video that's pure gold: how to write prompts that deliver ridiculously good results. I've seen $300 courses that don't even come close to what he explains in the first 10 minutes. CLAUDE.md files, memory shortcuts, parallel sessions, and prompting patterns that almost no one uses... All in one single video. Completely free. No fluff. Whether you're a developer, just starting, or have already been using Claude for months, this changes the game for you starting today. Save. Watch. Share. https://t.co/0PdoWJfTSr
29.8K4:00Claude Code built roughly 95% of this procedural pixel creature engine from a single detailed prompt. No premade sprites. No AI-generated textures. Just Godot, C#, and code generating body plans, rigs, animations, mutations, and pixel art across 9 creature families. The prompt wasn’t a vibe. It was an engineering spec. Check the repo in the comments.
172.9K0:25Official Anthropic tip for Claude Code: stop burning Opus 5.5 context on tasks Sonnet 5.5 can swarm, while Fable 5.1 sits idle spin up an autonomous team with /teams claude --teammates ux,backend,adversary Opus 5.5 acts as the lead architect, scoping architecture and validating pull requests Sonnet 5.5 teammates take isolated git worktrees and hammer out implementations at medium effort Fable 5.1 monitors the peer-to-peer bus as a dedicated adversary, only stepping in at three critical moments: → before a contract locks: does the frontend payload match the backend schema? → when a test breaks twice: are we patching the bug or just hiding the symptom? → before calling done: what attack surface or edge case did everyone overlook? Sonnet 5.5 builds. Fable 5.1 challenges. Opus 5.5 merges and ships Jev engineering runs the identical pattern one layer down: mechanical decisions that need no reasoning (which file to open, which tool to invoke, retry or abort) execute in 16ms, so the fro
66.3K26:36مهندس من Anthropic كشف عن طريقة لتشغيل أكثر من 100 وكيل Claude Code يعملون كفريق واحد، بإدارة وكيل رئيسي يتولى التخطيط والتنفيذ. وفي ورشة مدتها 30 دقيقة، شرح كيف تستفيد من Claude Code بأقصى إمكانياته وأقل تكلفة. محتوى يستاهل وقتك إذا تبي تحوّل الذكاء الاصطناعي من مجرد أداة إلى فريق عمل متكامل. https://t.co/y7nSxOE88g
4.4M1:12You can now mod Claude Code: - Change how it behaves - Customize the UI - Swap in your own features Write one with a few lines of TypeScript, or have Claude build it for you. Mods ship inside plugins, so you install them with /plugin in the CLI or desktop app. A few examples: https://t.co/qMtzV8AeId
91.4K0:22I don't understand why everyone isn't using this yet this will make your Claude boost productivity by 300% Anthropic engineer Lance Martin revealed a skill for Opus 5.5 Open Claude Code and type: /claude-api prompt-audit > Audits your skills, CLAUDE.md and prompts > Removes everything that slows down the model > Rewrites using the official Opus 5.5 guide Your prompts are written for older models This can be fixed in 3 minutes Bookmark this
—0:38
98.4K0:071/ Excited to release The ultimate guide to multi-harness RL The same model behaves differently in every agent harness. So we built an open way to train any model with RL on any task set, inside the harnesses people actually use, like Claude Code, Codex, and OpenCode, without changing a single line of harness or training code. Trained across four harnesses, LFM2.5-2.6B went from 42% to 54% with 31% fewer tool calls. 🧵
About AI Coding launch videos
Collected from XEvery entry is a public X post from this niche that carries a native video. We keep the post text, author, poster frame and live counters, and link to the post on X rather than re-hosting the video. This is the same sampling described in the methodology: a high-quality sample of what shipped, not a census of X.