Posts with #Codux
#Codux
Public posts on X that use #Codux.
one year at Stanford costs more than $60,000 the CEO of Obsidian put his second brain setup on GitHub for $0 Steph Ango shared his vault, his theme and his agent skills: 7 repos, 77,106 stars. the vault template alone is a f*cking masterclass • layer 1: the vault > kepano-obsidian - his personal vault template, organized bottom-up ↳ > 40-questions - what he asks himself at the end of every year and every decade ↳ • layer 2: getting the web in > defuddle - any web page as clean Markdown, built for the Obsidian Web Clipper ↳ > clipper-templates - ready Web Clipper templates for popular sites ↳ • layer 3: handing it to an agent > obsidian-skills - teaches Claude Code and Codex the Obsidian CLI and its open formats ↳ • layer 4: making it yours > obsidian-minimal - the distraction-free theme, tunable to the last detail ↳ > flexoki - an inky color scheme for prose and code ↳ copy layer 1 before layer 3. an agent is only as useful as the vault you point it at Karpathy's take on the same idea, Claude skills + Obsidian, is in the article below ↓Mr. Buzzoni @Mr_Buzzoni144035324.4K
GITHUB IS ACTUALLY INSANE RIGHT NOW. You can build an entire AI content factory without hiring a team. Here are 10 open-source repos worth checking out: 1. OpenCreator Give Codex a creator workspace for writing scripts, generating images, creating voiceovers, translating videos, and editing content. 2. Social Media Research Skills Use Claude Code or Codex to research viral posts, analyze competitors, mine comments, and extract hooks from video transcripts. 3. Local Deep Research Turn any topic into a detailed research report using web searches, your documents, and local or cloud AI models. 4. FLUX.2 Generate and edit images for thumbnails, social posts, and marketing campaigns. The Klein 4B models have Apache-licensed weights. 5. Kokoro Turn written scripts into AI voiceovers with a compact, open-weight text-to-speech model. 6. ACE-Step 1.5 Generate music from text prompts, create instrumentals, and regenerate specific audio sections. It can run locally on supported hardware. 7. Short Video Factory Create batches of short videos from prompts and your footage, with AI-generated scripts, voiceovers, editing, and subtitles. 8. AI YouTube Shorts Generator Convert long YouTube videos into vertical Shorts with AI-selected highlights, transcripts, and automatic cropping. 9. Podcastfy Turn source material into AI-generated podcast conversations, customize the dialogue, and produce audio in multiple languages. 10. Social Media Agent Turn a URL into X and LinkedIn posts, review and edit the drafts, then schedule them through an AI agent workflow. Research, writing, visuals, voiceovers, music, video editing, and distribution. You can automate almost every stage of content creation with the right tools. Which part of your workflow would you automate first? Want to build your own AI content factory? 👀 Follow @rakib_hossen_ai for more AI tools, open-source gems, and workflows that help you create smarter and faster. 🚀
408240422.1K
Hi guys, I rebuilt my portfolio with three AI agents (Claude Code, Cursor and Codex). Real case studies with the numbers to back them, a sword game with a live leaderboard, and a dragon that tells you facts about me. Have a look:1781801.2K120K
卧槽！！！Codex 省 token 神器来了！ GitHub 11.5 万+ Star！AI 输出 token 节省 65%！ 以前用 Codex，修一个 Bug 都能输出几百行废话，Token 烧得飞快 现在装上 Caveman，直接让 AI 进入「原始人模式」！ 【完整安装教程看视频，一定要用这段实测过的提示词】 最核心的价值： 1、最多减少 65% 输出 Token：精简 AI 的表达，减少冗长解释。 2、减少上下文噪音：压缩不必要的信息，让关键信息更突出。 3、适配 Codex：通过项目提供的配置方式接入现有开发工作流。 求你们了！别再让 AI 一边写代码，一边疯狂烧 Token 了！govin.eth | G哥 @goan999999GPT-6.1 Sol + 6.1 Terra 组合，Token 消耗目标直降 80%！ 昨天刚发布的Terra一定要合理利用起来，别再让最强模型从头干到尾了 我给 AI 编程助手设计了一套「动态模型路由」策略：简单任务用 Terra，复杂难题用 Sol，解决后再切回 Terra 3 条规则，把模型额度花在刀刃上： 1、日常任务交给 Terra：翻译、摘要、文案、常规代码修改，能直接完成就不升级，减少不必要的高阶模型调用 2、复杂难题升级 Sol：遇到架构设计、跨模块 Bug、复杂推理，或者 Terra 两次修复仍未通过，直接升级，避免反复试错 3、难题解决立即降级：Sol 负责定位根因、解决关键问题，Terra 接手测试、常规修改和最终交付 核心逻辑就一句话： 简单任务不浪费强模型，复杂任务不让轻模型硬扛！ 完整动态模型路由规则放在评论区，直接复制给你的 Codex 编程助手！👇173728256.2K- GPT-6.1 Sol + 6.1 Terra 组合，Token 消耗目标直降 80%！ 昨天刚发布的Terra一定要合理利用起来，别再让最强模型从头干到尾了 我给 AI 编程助手设计了一套「动态模型路由」策略：简单任务用 Terra，复杂难题用 Sol，解决后再切回 Terra 3 条规则，把模型额度花在刀刃上： 1、日常任务交给 Terra：翻译、摘要、文案、常规代码修改，能直接完成就不升级，减少不必要的高阶模型调用 2、复杂难题升级 Sol：遇到架构设计、跨模块 Bug、复杂推理，或者 Terra 两次修复仍未通过，直接升级，避免反复试错 3、难题解决立即降级：Sol 负责定位根因、解决关键问题，Terra 接手测试、常规修改和最终交付 核心逻辑就一句话： 简单任务不浪费强模型，复杂任务不让轻模型硬扛！ 完整动态模型路由规则放在评论区，直接复制给你的 Codex 编程助手！👇
421113262.6K 
One month left to use your ChatGPT quota inside Codex before the limits merge. My tool lets you run Web models, including Pro, directly in Codex without spending Codex quota. If you missed it, now's a good time to try it. Repo in replies.Token Gremlin @TokenGremlinBad news for ChatGPT users: starting in December, Plus, Pro and Business limits will be shared across ChatGPT and Codex. OpenAI quietly updated its help page. Usage will now be measured in tokens, so bigger models and harder tasks burn through your allowance faster. Only Instant chat stays free of the limit. Hit the cap and you wait, upgrade or buy credits. Heavy Codex users may feel this one.15515324.5K
Bad news for ChatGPT users: starting in December, Plus, Pro and Business limits will be shared across ChatGPT and Codex. OpenAI quietly updated its help page. Usage will now be measured in tokens, so bigger models and harder tasks burn through your allowance faster. Only Instant chat stays free of the limit. Hit the cap and you wait, upgrade or buy credits. Heavy Codex users may feel this one.
3011211.7K283K
DEJA DE MONTAR TU SETUP DE CLAUDE CODE DESDE CERO Aquí te reuní 10 repositorios de GitHub que ya lo hicieron por ti: guías, skills, agentes, hooks y reglas listas para copiar REPOSITORIOS DE GITHUB: 1. Buenas prácticas de Claude Code → la guía completa de subagentes, skills, hooks, MCP y memoria, con ejemplos → 2. Claude Skills → 388 skills listas en 20 áreas, para Claude Code, Codex y Gemini CLI → 3. Kit de herramientas para Claude Code → 135 agentes, 42 comandos, 20 hooks y 15 reglas para meter en tu proyecto → 4. Kit para crear productos con IA → 6 prompts de planificación, una plantilla de Next.js y un CLAUDE.md listo → 5. Skills de vídeo para Claude → más de 300 repos que enseñan a Claude a hacer vídeo, ordenados por tipo → 6. Awesome Claude Code → el mapa más limpio de herramientas, plugins y contadores de uso → 7. Skills de motion graphics → animaciones tipo lanzamiento hechas con código, sin After Effects → 8. Reglas para tareas largas → define qué es "terminado", lleva un TASKS.md y cierra con un informe fijo → 9. Model Bump → revisa tu código y te dice qué se rompe al pasar a Opus 5.5 → 10. Claude Best Practices → 29 agentes, 174 skills, 58 reglas y 22 hooks para copiar → Lo que importa no es cuántas estrellas tiene un repo, sino que te funcione. Elige uno, pruébalo en una sola tarea y quédate solo con lo que sirva.ismael. @n0tisma371541.1K74.3K
🚨 GitHub تحلّ رسمياً أكبر مشكلة في الـ Vibe Coding! أطلقت GitHub أداة جديدة باسم Spec Kit، وتجاوزت 120 ألف نجمة على GitHub في وقت قياسي جداً! 🔥🧵 بدلاً من كتابة برومبت عشوائي والدعاء ألا يدمر الذكاء الاصطناعي مشروعك.. Spec Kit تجبر الـ AI على كتابة مواصفات وهيكلة منتظمة قبل لمس سطر كود واحد! كيف يعمل النموذج؟ 1️⃣ يفهم ما تريد بناءه بدقة 2️⃣ يطرح عليك الأسئلة والمعطيات الناقصة 3️⃣ يهيكل المشروع بالكامل 4️⃣ ثم يبدأ في كتابة الكود! النتيجة؟ ساعات أقل في إصلاح الأخطاء العشوائية، كود نظيف ومنظم، ونتائج مضمونة عند التعامل مع الوكلاء الذكيين (Agents). 🔄 سلسلة أوامر العمل بسيطة ومباشرة: • constitution/ 👈 القواعد والمعايير الأساسية • specify/ 👈 ما الذي تريد بناءه بالضبط؟ • clarify/ 👈 أسئلة وتوضيحات قبل البدء • plan/ 👈 المعمارية والتقنيات المستخدمة (Stack) • tasks/ 👈 تقسيم المهام بالترتيب • implement/ 👈 مرحلة التنفيذ والتكويد 💡 المفاجأة: الأداة متوافقة ومجربة مع أكثر من 25 وكيل ذكي (مثل Cursor, Claude Code, Copilot, Gemini CLI, Codex وغيرهم). • 🌟 أكثر من 120,000 Star • 🍴 أكثر من 10,000 Fork 🔓 مفتوحة المصدر بالكامل من GitHub 📌 احفظ التغريدة للرجوع إليها وقت الحاجة 🔖 رابط المستودع (Repo) تجده في أول تعليق 👇343022.7K235K
🚨Codex Users!!! Codex limits are getting cut in half tomorrow, and I genuinely don't understand the logic behind paying $200 for a 10x OpenAI plan. Astra 6 and Sol 6.1 just don't justify that price for power users. Who thought this was a good move? Because Anthropic looks like the biggest winner here. Opus 5.5 is becoming the only serious option for people paying $200 for a 20x plan. And as more power users move to Claude Code, Anthropic gets more usage, more feedback, and more real-world coding data to improve its next models. OpenAI might be handing its biggest advantage to its competitor.188942K454K- 🚨 OpenAI could be making ChatGPT Plus a much harder subscription to justify. Starting in December, Plus, Pro, and Business users will have one shared usage allowance across ChatGPT and Codex. Instant chat is supposed to remain free of usage charges, but the bigger concern is how much room will be left for everything else. For Plus users, this could mean choosing between using ChatGPT and getting work done in Codex with the same limited allowance. If OpenAI gets the balance wrong, Anthropic might gain a lot of frustrated customers. I really hope they rethink this before December.
985766582.2K 
You asked, I shipped The unslop skill now works in Codex tooDamir Sokolovsky @heydamirI turned my unslopping process into a free Claude skill It works on landing pages, dashboards and mobile apps Go unslop your stuff:131237425.3K- I turned my unslopping process into a free Claude skill It works on landing pages, dashboards and mobile apps Go unslop your stuff:1342285K565K
Switched from Codex to Claude Code on the same $20 plan, and I'm genuinely blown away. I've been using Codex for a while, but recently decided to give Claude Code a shot. Same $20/month, but the difference feels massive. It genuinely feels like I'm using AI from a year in the future. I was expecting Claude to burn through my quota pretty quickly, especially with Opus 5.5. But surprisingly, I'm struggling to even exhaust it. Here's a real example: I gave Sol 6.1 a Figma design to implement, including proper testing across the specified breakpoints. It burned through the entire 5-hour quota and only managed to complete around 70% of the work. Then I tried the same task with Opus 5.5 on medium effort. It completed the whole thing, delivered noticeably better quality, and used only about 50% of my 5-hour quota. I'm genuinely confused. Did Anthropic accidentally give me the Max plan or something? Because this is ridiculous. I know it's just one comparison, but the difference in my experience has been insane. To all the Codex users: seriously, give Claude Code a try. You might be surprised how far behind you've been feeling without even realizing it.57821129.1K
Introducing @computer. Claude Code & Codex in the cloud, as fast as your local machine. Sometimes even faster! Try it today via Slack, byo sub supported.413324541.1K
holy f*ck this repo lets you make your own pokemon: one picture in, a full game-ready sprite set out agent-sprite-forge, a skill for Claude, Codex and Grok: > approve one master image and get a clip for every action: idle, walk, run, attack, jump, hurt > python scripts cut the background, align the frames, loop them and fix the timing > pixel mode keeps a limited palette, colors stay true to your master image > export to a PNG atlas, WebM with alpha, Godot SpriteFrames or Aseprite > generate2dmap builds tiles and whole levels for Tiled, Godot and LDtk > every call is logged with its cost, plus a budget cap so it never runs away 4,700+ stars on github my PokeTide's pokemon crew approves xdMnimiy @Mnilaxit's f*cking gold the physics engine behind Death Stranding 2 is free on github i used these to build my own game with 20,000+ players 10 github repos that turn Opus 5.5 into a full AAA game studio: MAKE IT MOVE LIKE AAA 01. Jolt Physics: -> the engine behind Horizon Forbidden West and Death Stranding 2: rigid bodies, collisions, real physics 02. ozz-animation: -> load, play and blend character animations, built for speed 03. Recast Navigation: -> navmeshes, pathfinding and crowds that walk around each other BUILD HUGE WORLDS 04. Terrain3D: -> sculpt mountains, paint the ground and grow foliage across giant Godot maps 05. Voxel Tools for Godot: -> worlds players can dig and blow up: caves, tunnels, overhangs, streaming as you explore 06. FastNoise2: -> mix noise in a visual node editor, drop the result straight into your game MAKE IT LOOK EXPENSIVE 07. Material Maker: -> procedural textures from nodes, plus painting right on your 3D models 08. Effekseer: -> explosions, magic and combat effects, exported as 2D or real 3D SHIP YOUR OWN ENGINE 09. The Forge: -> cross-platform graphics, asset loading and animation, so you don't start from zero 10. meshoptimizer: -> simplify and compress meshes for faster rendering on heavy scenes every system of a AAA game, one link each what should i build with them next? Pokemon Unite v2?185263877.3K- it's f*cking gold the physics engine behind Death Stranding 2 is free on github i used these to build my own game with 20,000+ players 10 github repos that turn Opus 5.5 into a full AAA game studio: MAKE IT MOVE LIKE AAA 01. Jolt Physics: -> the engine behind Horizon Forbidden West and Death Stranding 2: rigid bodies, collisions, real physics 02. ozz-animation: -> load, play and blend character animations, built for speed 03. Recast Navigation: -> navmeshes, pathfinding and crowds that walk around each other BUILD HUGE WORLDS 04. Terrain3D: -> sculpt mountains, paint the ground and grow foliage across giant Godot maps 05. Voxel Tools for Godot: -> worlds players can dig and blow up: caves, tunnels, overhangs, streaming as you explore 06. FastNoise2: -> mix noise in a visual node editor, drop the result straight into your game MAKE IT LOOK EXPENSIVE 07. Material Maker: -> procedural textures from nodes, plus painting right on your 3D models 08. Effekseer: -> explosions, magic and combat effects, exported as 2D or real 3D SHIP YOUR OWN ENGINE 09. The Forge: -> cross-platform graphics, asset loading and animation, so you don't start from zero 10. meshoptimizer: -> simplify and compress meshes for faster rendering on heavy scenes every system of a AAA game, one link each what should i build with them next? Pokemon Unite v2?47485763.2K
NOBODY SHOULD BE BUILDING A CLAUDE CODE TEAM FROM SCRATCH ANYMORE Anthropic made Agent Skills an open spec, and GitHub filled up with ready-made Claude Code teams you can install for free one repo alone claims 135 agents, 35 skills, 42 commands, 176+ plugins and 20 hooks. another hands you 100+ specialized subagents. none of it costs a cent 10 GitHub projects worth opening first: 01 wshobson/agents → a plugin marketplace that installs the same agents into Claude Code, Codex, Cursor and Copilot, so your setup survives a tool switch 02 VoltAgent/awesome-claude-code-subagents → 100+ specialized subagents for most of what a normal engineering week throws at you 03 rohitg00/awesome-claude-code-toolkit → agents, curated skills, commands, plugins, hooks and MCP configs in one place 04 anthropics/skills → the official Agent Skills repo. look at a working SKILL.md here before you copy anyone else's 05 netresearch/claude-code-marketplace → curated skills on an open standard, portable across Claude Code, Cursor, Copilot, Codex and Gemini CLI 06 disler/claude-code-hooks-mastery → the hooks reference. learn to make the agent stop on your rules instead of stopping to ask you 07 karanb192/claude-code-hooks → an installable hooks marketplace split into safety, cost, observability and productivity 08 affaan-m/claude-swarm → multi-agent orchestration that splits the task, runs the agents and shows the whole run in a terminal UI 09 am-will/swarms → orchestration shipped as skills, so the coordination only loads when the job calls for it 10 eugeniughelbur/obsidian-second-brain → memory as plain markdown in an Obsidian vault, shared across Claude Code, Codex and other agents give your first team a job with an obvious finish line: one reviewed PR, one researched brief, one failing test turned green then check how much of that output you actually kept and what it cost. a bigger roster only earns its place when it moves that number the download is free. paying four agents to redo the same investigation is not save this before you burn another weekend building a config that already existseigen @e1genqq432515210.7K
Harness engineering is the most f*cking DANGEROUS thing to understand and nobody is learning it like this researchers broke down Claude Code, Codex, and 9 more AI agents to reveal how coding agents are actually built the paper maps the systems into 7 parts - and includes a 90-line starter harness you can study and adapt. here's the blueprint ↓ step 1 → build the agent loop. send the task to the model, execute its actions, return the results, and repeat until it finishes or hits a limit step 2 → connect the model. assemble the instructions and handle the provider's response and tool-call format step 3 → give it tools. start with what the task needs: search files, read code, edit files and run commands step 4 → manage context. load relevant project instructions, keep useful results, and compact history when it gets too long step 5 → define permissions. decide what runs automatically, what needs approval, and what should be blocked or sandboxed step 6 → add orchestration when the work needs it. a single agent can start simple; sub-agents can split and coordinate independent tasks step 7 → make it extensible. hooks, skills and plugins let you add behavior without rebuilding the whole runtime one finding: in the 11 systems studied, the authors found no general-purpose agent framework imported in the agent runtime and no vector-based code retrieval the practical part is section 16: 18 design recommendations and a starter implementation. the 90-line scaffold is illustrative and omits a sandbox, so it needs more work before real deployment save this, then read the full guide for Opus 5.5 Harness Engineering ↓
beamnxw ./ @beamnxwStop letting unfinished Opus 5.5 runs eat your budget Set up the harness before your next long task Save progress, verify results, reuse the workflow The full 7-layer setup below ↓545150935.3K- Stop letting unfinished Opus 5.5 runs eat your budget Set up the harness before your next long task Save progress, verify results, reuse the workflow The full 7-layer setup below ↓821891.5K2.5M
I am hearing that most developers use the Desktop apps for Claude Code and ChatGPT Codex, not the CLI. That can't be right… can it?311211.2K413K- Show more posts
