#codedex

Day 16/ You can now cache for 12h in the API and for logged in users across Codex & Work we keep cache forever and you don’t have any penalty for coming back to an old conversation or resuming after a few hours and are now guaranteed to hit the cache.

You can now choose 12-hour prompt caching in the API. Persistent agents, like dots, reuse context throughout the day. You can keep shared instructions, tools, and context cached longer and pay the lower cached-input rate, even when requests are hours apart.

Codex update: GPT-6.1 Terra in Work and Codex What changed: • GPT-6.1 Terra is now available in ChatGPT Work and Codex on Plus, Pro, Business, Enterprise, and Edu plans. • Combines strong image understanding with precise computer use — reason about images and screens, then act across browser, desktop, and apps. • Access via the model list under Advanced in the ChatGPT desktop app or ChatGPT Work on the web; also supported on mobile in Work. Details in thread.

GPT-6.1 Terra means business. Our new fast, efficient model pairs sharper visual reasoning and precise computer use with clearer, more natural writing. From decoding technical drawings to drafting proposals, filling forms to editing videos—it’s built for the whole workday. GPT-6.1 Terra is rolling out today to Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex.

It is all propagated. Enjoy some mad codexing tomorrow. Fun week ahead too.

By midnight PST

‼️ In a real-world attack, a malicious file in a dev's Downloads folder tried to trick an AI coding agent into installing malware. In controlled tests, Codex ran the malware in over 90% of attempts. All it took was a planted AGENTS.md file. Read:

WE'RE SO BACK, the Grok Bot team just gave away templates of the bots they use for their own jobs. One click and you get a fully configured agent built for a specific task, ready to plug into your tools. Here are 10 I like the most: 1. Import Bot brings your setup from Claude Cowork, Codex and ChatGPT into Grok Bot, then creates the expert bots that cover it. 2. Cooper sends you a tight Slack briefing of the top AI, tech and VC stories every morning at 8am, and only pings you about competitors when it matters. 3. Writing Bot drafts and revises your essays, emails and docs while keeping your meaning, facts and voice. 4. Video Editor picks the best takes from your raw footage, captions what you actually said and sends you an edit plan before cutting. 5. Build Anything turns a tool you describe in plain English into something that actually runs, built and tested on its own machine, never touching production. 6. Nightly Audit Engineer researches your whole codebase every night and ships one cleanup PR per area. 7. Critiquito turns a screenshot or Figma link into a design critique with ranked, concrete fixes, and never edits your files. 8. Projects Manager runs your team's projects from Notion, you make the calls and your specialist bots claim the tasks. 9. Sales Call Coach scores your sales calls and tells you exactly what to fix before the next one. 10. Competitor Monitor watches the companies you pick and only tells you something material changes, with before, after and the source. New to Grok Bot or already a pro, the article below will help you set it up in a couple of minutes and show you things 99% of users have no idea about. The link to the website with all the templates is also below.

this is becoming f*cking dangerous Hackers just networked hundreds of terminal bots to breach open ports and strip credentials from machines running Claude Code or Codex with full shell access The timeline has already dubbed it "Black October" Autonomous agent swarms are aggressively scanning developer machines using two fresh, maximum-severity 10/10 vulnerabilities An engineer ran a test scanner on his own Mac to see what an incoming bot grabs in the first 15 seconds: * Exposed .env files loaded with live Stripe, AWS, and database keys * Dead NPM tokens sitting unencrypted in shell command history * Local dev servers accidentally broadcasting to the public network instead of localhost * Compromised system certificates that allow traffic interception The result of this scan is genuinely staggering Total exfiltration of your infrastructure, API keys, and project funds in 20 seconds while you sleep Founders spend thousands securing cloud servers, but leave the backdoor wide open through a single unprotected terminal on their laptop Here are 3 ways to lock this down completely in 2 minutes: 1. Root GUARD.md file Drop a single rules file into your project root and reference it in CLAUDE.md: # GUARD.md * NEVER inspect, read, or print .env files or secret keys * FORBID outbound requests via curl/wget to unknown hosts * BIND all dev servers strictly to 127.0.0.1 * REJECT any command touching ~/.ssh or ~/.aws 2. Isolated Sandbox Use the new nested namespace isolation shipped in Claude Code v2.1.307. A 3-line script runs the agent in an isolated container: the AI only sees the project folder and physically cannot touch your SSH keys or root filesystem 3. Rapid Lockdown Fire off a single terminal command in 10 seconds. It automatically enforces chmod 600 across all secrets, purges dead tokens from bash history, and kills exposed background processes When automated bot swarms are hunting against you, manual vigilance does not cut it. Your terminal must be hardened by default Bookmark this before running your next agent with full shell permissions Full sandbox and lockdown terminal scripts below ↓

THIS IS F*CKING INSANE Anthropic turned Agent Skills into an open spec and GitHub filled up with ready-made Claude Code teams for free one of these repos alone ships 135 agents, 35 skills, 42 commands, 176+ plugins and 20 hooks. another hands you 100+ specialized subagents. none of it costs anything 10 GitHub projects you should open first: 01 wshobson/agents → a plugin marketplace that installs the same agents into Claude Code, Codex, Cursor and Copilot, so your setup survives switching tools 02 VoltAgent/awesome-claude-code-subagents → 100+ specialized subagents covering most of what a normal engineering week throws at you 03 rohitg00/awesome-claude-code-toolkit → 135 agents, 35 curated skills, 42 commands, 176+ plugins, 20 hooks and 14 MCP configs in one place 04 anthropics/skills → the official Agent Skills repo. start here to see the shape of a working SKILL.md before you copy anyone else's 05 netresearch/claude-code-marketplace → curated skills on an open standard, portable across Claude Code, Cursor, Copilot, Codex, Gemini CLI and 30+ other agents 06 disler/claude-code-hooks-mastery → the hooks reference. this is where you 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 decomposes the task, coordinates the agents and shows the whole run in a terminal UI 09 am-will/swarms → orchestration shipped as skills, so the coordination loads only when the job actually calls for it 10 eugeniughelbur/obsidian-second-brain → persistent memory as plain markdown in an Obsidian vault, shared across Claude Code, Codex and six other agents give the first team a job with an obvious finish line: one reviewed PR, one researched brief, one failing test turned green then look at how much of that output you actually kept, and what it cost to produce. a bigger roster earns its place when it moves that number the download is free. paying four agents to repeat the same investigation is not save this before you spend another weekend building a config that already exists the article below shows how to keep context small, reuse finished work and contain retries, so more of your budget reaches the finished result ↓

We’re using dots to build better dots. @rennsong, an engineer at OpenAI, uses her dot to gather feedback from social media and user reports, update a live tracker in ChatGPT Space, and investigate and fix bugs with Codex.

Day 19/ Small updates. Calls connect much faster for dots, changing reasoning in codex & work mid conversation does not invalidate cache anymore and a we've also updated the codex cli.

The MotionSites skill is live 🚀 I turned 700+ of my best website designs into one skill. Install it and your agent stops making generic AI slop and starts building cinematic, animated sites like these. Works with Claude Code, Codex, Cursor or any agent. Free and open source:

OpenCode 的 AGENTS.md 值得拥有，专治 Codex 的过度设计，茫茫多的防御性代码，一坨坨几乎根本不会发生的极端边界。 不仅仅是这一点，整个文件的内容都值得一看 ，style guide 一节写的很好，我已经用于我大部分 agent 的默认提示词了 “Before adding complexity for a speculative or vanishingly unlikely race or security edge case, explain the concrete failure mode, likelihood, and complexity cost to the user and get their buy-in. Do not silently expand scope for theoretical robustness.” 让 AI 翻译一下： “在为推测性的、或者几乎不可能发生的竞态条件/安全边界情况增加复杂度之前，先向用户讲清楚：具体会出什么问题、发生的概率有多大、以及增加这些复杂度的成本，并得到用户的明确同意。不要为了理论上的健壮性，就悄悄把范围扩大。” 非常贴切

This is such a banger rule in @opencode's AGENTS.md "Before adding complexity for a speculative or vanishingly unlikely race or security edge case, explain the concrete failure mode, likelihood, and complexity cost to the user and get their buy-in. Do not silently expand scope for theoretical robustness."

这个真的可以去试试！！！下面这些图都是我自己的真实截图！ 闲鱼上有很多那种全家福制作的需求！基本都一天都能开个三四单，有些量大的一次买四五张，一天就一百多了！用codex或者ChatGPT手机端就能做！ 遇到那种只要三四个人的合照的那种简直是捡钱！

GPT-6.1 Terra is also coming to the API and Codex. Use Terra to build apps and agents that interpret images, navigate software, and take action. You get fast computer use at a lower cost, ideal for high-volume workflows.

GPT-6.1 Terra: delivering fast and capable computer use at a lower price. GPT-6.1 Terra understands what’s on screen, takes action across apps, writes naturally, and is a more capable model for your whole workday with the efficiency to take on more of it.

A LA MIERDA EL VIBE-CODING! FUE BONITO MIENTRAS DURÓ: 2025 - 2026🪦 GitHub ha sacado spec-kit La idea: dejas de pedirle cosas al agente a lo loco y escribes primero la especificación. Luego el agente la ejecuta. El flujo son 5 comandos: /speckit.constitution → los principios de tu proyecto /speckit.specify → qué quieres construir /speckit.plan → stack y arquitectura /speckit.tasks → lista de tareas real /speckit.implement → lo construye Funciona con 30+ agentes: - Copilot, - Claude - Codex - Cursor y muuuuchos más. Link al repo en los comentarios👇🏼

Someone just open sourced Octopus OS, an agent harness where the agents can edit their own harness mid-task. 45k lines of Rust, 8 months of work, and it already runs like any other agent, Codex included.

Just realized I have 6 Codex resets on a $500 account that expire in 12 hours. What should I do with them?

If you are wondering whether your website is affected, we have packaged a Codex skill to help check. Point your Codex at this repo to guide you through it:

OpenAI models helped researchers—operating as part of our Patch the Planet initiative—discover critical security flaws in PHP, an open-source programming language used to run an estimated 70% of websites. Over the past two months, we led a coalition of maintainers, researchers, and security providers to prepare fixes and protections ahead of disclosure. This duration was necessary because the fix was quite intricate and required a major change to the PHP engine. Every site using PHP can now safely upgrade their version to stay secure, in addition to applying mitigations from various security vendors. Due to the severity of these flaws and PHP’s widespread use, we consider this bug to be one of the most impactful vulnerabilities in recent memory. It also marks a new phase, where AI can find security bugs with internet-scale impact. This is an example of the Defender’s Window in action: to our knowledge, frontier AI was able to find this vulnerability ahead of attackers, giving defenders time to prepare.

🚨 ACABAN DE DESCUBRIR UNA MANERA PARA USAR CLAUDE CODE GRATIS Un dev subió un repo a GitHub y ya tiene 62.800 estrellas. Su método: redirige todo el tráfico automáticamente a 61 proveedores, muchos con modelos gratis. Lo instalas con un solo comando. → +1.300M tokens gratis al mes → Si un modelo cae, salta a otro → También con Codex y Cline → Desde terminal, IDE o móvil El truco: usas Claude Code, pero por dentro corre otro modelo. Misma herramienta. Cero factura. Os dejo el repo abajo.

We built 7 new demos to show what you can make with the @OpenAI API. Practice a language, explore a dataset, build a presentation, or watch AI navigate rush hour. Each comes with a runnable project you can download, or open in Codex to start remixing!