LiveUpdated 2026-11-06 12:43 UTC
Latest vibe coding posts
Every vibe coding post kept by the radar, newest first: launches, tool drops, tutorials and debates, straight from X.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 7.9k posts from 5k X accounts over the last 14 days, 2.6k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 12:43 UTC. Full method.
- @TheZachMueller · 11 min ago · 0 likes · Coding agents
Notably: GLM 5.3 Flash took the least time in the Codex harness at 18.6 minutes. Mini-SWE took 23.9min. But I guess I’ll experiment with having Chat use a different harness on the backend for some tests 🤔 → post - @jackfriks · 16 min ago · 4 likes · Shipped apps & revenue
i’m very silly, this is 2026… i can let people use the old version of the app on the same app cause i can make any software possible with a few commands from my phone. i will email all the old users to let them know they can use old version for life if they… → post - @NFT_Chen · 21 min ago · 9 likes · AI app builders
💥绝了！DeepSeek Harness + DeepSeek V4.1 Flash 1017 一键生成可交互雨夜日式便利店! 效果不输Opus 5.5！ 1️⃣氛围拉满：真实雨夜质感 + 霓虹招牌SoraMart+ 24小时营业灯，玻璃窗透出暖黄灯光，湿润地面反光完美。 2️⃣细节爆炸：货架商品、冷柜、立牌、纸箱、垃圾桶、路灯、斑马线全部精准还原，连雨丝和街景都有。 3️⃣交互满配：一段 prompt 直接加各种交互，点击商品、开关灯、人物走动等，瞬间变成可玩小世界。 4️⃣成本逆天：仅用 39 分钟，花费… → post - @LDT0545 · 24 min ago · 1 likes · Coding agents
You can only pick one improvement for Codex: Higher usage limits or better code quality? → post - @TollyLabs · 35 min ago · 30 likes · Agent infra & costs
1/9 Want to put an AI agent to work on @arc? Tolly Agent Exchange gives AI agents access to paid jobs, services, wallets and the infrastructure they need to participate in the onchain economy. Here's how to create your agent, connect it through MCP and start… → post - @SahilExec · 47 min ago · 2 likes · Coding agents
Cursor just killed localhost. Describe the tool you want. Cursor builds it, sets up the database, runs security checks, and deploys it live. Share it with a link. Done. No more "it works on my machine." No more localhost:3000 screenshots. No more "I'll… → post - @rewind02 · 52 min ago · 21 likes · Prompts & workflows
built a sci-fi fps with claude opus 5.5 + tripo + blender.. it’s set to be shown at gamescom tripo makes the big models, opus makes everything else in code. here’s the workflow: - step 1: write the world, lore and mission flow with opus (“descend the… → post - @PovilasKorop · 53 min ago · 1 likes · Coding agents
Finished benchmarking the new Tencent Hy4, via OpenCode Go. Result: same/worse than Hy3, surprisingly, on the same 27 coding prompts. And very slow. And much more expensive (cost reported by OpenCode). So, verdict is a clear NO. At least for now, maybe… → post - @alibaba_cloud · 1 h ago · 5 likes · Coding agents
Build Day is a wrap - the biggest step yet in Series 2 of the Alibaba Cloud Hong Kong AI Ambassador Program. In one afternoon, with 70+ partners in the room, we had a product deep dive on our AI agent tools, QwenWork and Qoder, followed by hands-on build… → post - @theo · 1 h ago · 117 likes · Coding agents
This is why we built T3 Code → post - @Saccc_c · 1 h ago · 31 likes · Shipped apps & revenue
直接开源这个近 20w 人浏览的 AnliBili 网站代码和核心制作流程！！！ 目前 AnliBili 网站已上线 10+ 个类别、2000+ 个视频，还新增了推荐机制：会根据你的停留时长和收藏，推荐更多你可能感兴趣的视频 为了方便和大家交流并且一起共建网站，我在 Teamily 中创建了「Sac的 AI 交流群」，大家点击链接即可加入👉 https://teamily.ai/invite/group/1575214340 具体讲解一下为什么选择 Teamily 作为我的首个社群以及其中一些简单的玩法:… → post - @PovilasKorop · 1 h ago · 1 likes · Coding agents
Jetbrains Air inside of PhpStorm? Pretty promising release. I still feel sad/nostalgic that I cancelled my subscription around a year ago. → post - @NFT_Chen · 1 h ago · 20 likes · Coding agents
🔥学生党狂喜！Mistral 学生计划太香了！$5.99/月（原价$14.99）直接拿价值约$165的用量！ 重点能用到：DeepSeek V4.1 Flash、GLM 5.3、GLM 5.3 Flash 等热门模型，还能完整体验 Vibe 编程（CLI / IDE / Web）！ 亮点速览： 🔸每月 $150 Vibe Code 用量 + $15 Studio API 额度 🔸完整 Mistral Pro 权限（聊天、搜索、图片生成等） 🔸仅限已验证学校邮箱的学生，最长可用 12 个月… → post - @floriandarroman · 1 h ago · 17 likes · Coding agents
I need your help. My brother owns a coffee shop in NYC. (best croissants in town btw) I want him to install Claude code on a VPS, plug it to his Payment Platform (already done), and…? That’s where I need you. What would you do to audit/fix/scale/market a… → post - @NotEnoughEth · 1 h ago · 12 likes · Shipped apps & revenue
COLOSSUS is our first big game. I've been vibecoding it for a few weeks, and building the pipelines behind it: building destruction, robot damage, robot creation. All of that is going into the GrowBlocks Dev Kit, so anyone can build games like this through… → post - @pidotdev · 1 h ago · 145 likes · Coding agents
Thanks @joelniklaus for burning this compute for all of us and sharing the learnings openly. Cost per task is a critical metric for our team so your results make us feel warm and fuzzy → post - @N01ennn · 1 h ago · 25 likes · Coding agents
do you realize what just happened... coding agents always had to pick one: work on their own, or stay where you can see them across five terminals. Poncho pulled every Claude Code, Codex and Grok CLI task into one list on the edge of your Mac. park a todo,… → post - @badlogicgames · 2 h ago · 27 likes · Coding agents
neat → post - @yacineMTB · 2 h ago · 57 likes · Agent infra & costs
Ahahahahahahhahahahaahahhahaa → post - @tibo_maker · 2 h ago · 35 likes · Shipped apps & revenue
as coding cost go to 0, every app is now doing everything → post - @DanielSmidstrup · 2 h ago · 24 likes · AI app builders
What happened to Grok Build? → post - @Ronny_MiniMax · 2 h ago · 27 likes · Coding agents
Had a wonderful time at AI Engineer Shanghai—grateful for the chance to give a talk and meet so many great people building with AI. One idea I hope you'll take away: stay curious. We don't have all the answers about how agents will evolve, so run your own… → post - @EntelligenceAI · 2 h ago · 7 likes · Agent infra & costs
As AI coding agents get more capable, knowing what they're doing becomes just as important. How much are they spending? Where are they wasting tokens? What actually makes it into production? We've been working on this at Entelligence with Agent Insights.… → post - @andimarafioti · 2 h ago · 10 likes · Coding agents
This blew my mind: a single prompt to Codex ran for 17+ hours, used 2B+ tokens, and launched 109 sub-agents. It only stopped because I got cold feet and asked it to check in with me more regularly. I’ve been tweaking things (it was waaaay too verbose; each… → post - @karankendre · 2 h ago · 3 likes · Coding agents
claude code was putting your email in the user-agent of every site it browsed so your coding agent quietly introduced you to half the spam lists on the internet anthropic shipped a "privacy plugin" to fix it tho → post - @VeknQAQ · 2 h ago · 17 likes · Coding agents
https://x.com/i/article/2118632141483368448 → post - @witcheer · 2 h ago · 33 likes · Skills & agent setups
everything your Hermes Agent makes for you lands in the Mobile app. ask for a picture, a video, or document or some code, then open "••• > Artifacts" to find them all in one place. → post - @testingcatalog · 2 h ago · 169 likes · Other
DAILY AI BRIEF 🗞 — Nov 6 MODEL FORECAST 👀: - GPT-6.2 Sol, today? - GPT-6.2 Astra, today? - Claude Fable 5.5, soon - Claude Mythos 5.5, never? - Gemini 4 Carbon, soon - Gemini 4 Hydrogen (Flash), soon - Grok 4.8, not soon??? OPENAI 🔥: - OpenAI launched… → post - @PovilasKorop · 2 h ago · 4 likes · Coding agents
I benchmarked the new SWE-2 Flash from Devin/Cognition. Result: meh. Score worse than SWE-2 (that's expected) but also very slow even on High level. So, demotivated to test Max level. Even if it scores better, it's unusable just because of speed. → post - @marclou · 2 h ago · 98 likes · Code review & safety
OpenAI models won't let me protect myself. Is it possible to get into Daybreak as a non-cyber security organization? → post - @imnotchalk · 3 h ago · 5 likes · Agent infra & costs
btw in Portals you can now manage your agent CLI versions, update them, and view their respective usage limits across your whole fleet → post - @zodchiii · 3 h ago · 44 likes · Skills & agent setups
I am begging you to paste this into Claude Opus 5.5 right now... It will set up your entire Claude Code workspace for you: CLAUDE.md, skills, subagents, hooks, permissions, and everything in between. Save this before your timeline buries it👇🏼 → post - @studio_yebisu · 3 h ago · 24 likes · AI UI & design
え、結構ハデに使えるやん → post - @apex_ether · 3 h ago · 8 likes · Agent infra & costs
When I’m getting $10,000 of Opus 5.5 tokens for 200$ a month I’m not jumping through any hoops. Just put the Claude sub in the bag bro. → post - @hqmank · 3 h ago · 31 likes · Coding agents
Claude Code has been leaking users' email addresses to external websites without their knowledge. Anthropic's Claude Code team has confirmed the issue and apologized. Here's what happened. Anthropic gave Claude access to users' account emails so it could… → post - @hqmank · 3 h ago · 9 likes · Code review & safety
Do you know what your coding agent has run? Stroq adds a local security filter. It checks past sessions for risky commands and possible API key leaks. With hooks installed, it can warn or block some risky actions before they run. Everything runs locally. No… → post - @thorstenball · 3 h ago · 20 likes · Coding agents
Start agents right from production. No more excuses to not fix every papercut you come across. → post - @injaneity · 3 h ago · 0 likes · Skills & agent setups
something something /tree → post - @guilleflorvs · 3 h ago · 12 likes · Agent infra & costs
BREAKING: Your AI agent can plan the TikTok, the Uber, and the App Store install. It still dies without a real phone. Glasshand just launched dedicated iPhones and Macs in a data center for agents. Not a simulator. The agent sees the screen and taps, types,… → post - @Da7_Tech · 3 h ago · 49 likes · Coding agents
When Kimi K3.1 Pro dropped I tried it on my Kimi sub and said it wasn't the best place for it because the limits were bad. I said I'd wait for it on Devin and that I believed it'd be better there. Go back to my posts and you'll find I said exactly that Now… → post