有没有发现，Codex 的额度越来越不经用了
本文来自微信公众号： MacTalk ，作者：池建强先说我自己的情况，我现在订的是GPT...
用 Seedance 2.5 拍《沙丘 3》，我们替你把 AI 短片的坑全都踩过了
看到这条视频，你会不会以为《沙丘 3》提前泄露了。 这其实是一部由我们团队的两位同事，花 12 天时间，结合 Codex 和 Seedance 2.5 完成的 AI 短片。 制作过程比我们预想中要复杂，整部片子只有 6 分钟，制作过程累计消耗了 3000 多元的积分额度。更重要的是，它并不是靠一句提示词「一键生成」出来的。...
福利：复旦「学术版 Codex」客户端上线，下载登录就送1 万积分，还可享会员 8 折
程序员被 Codex 彻底改变了，不用一行行敲代码，把需求丢给 AI，让它自己写代码、跑测试、修 Bug、闭环交付。 那么问题来了：科研人员天天干的活，不也是”配环境、跑实验、复现代码、查文献、写综述”这些体力活吗？ 现在，复旦大学 NLP 团队把这套逻辑搬到了科研上，做出了一个真正意义上的学术版 Codex切问学术。...
Coop – Isolated VM Environments for Running Claude Code and Codex
Article URL: https://github.com/trailofbits/coop Comments URL: https://news.ycombinator.com/item?id=49593842 Points: 7 # Comments: 1
WinToastRelay – 将 Windows 系统通知转发到手机｜Codex 跑完，手机会叫你
在 Windows 用 Codex 时，会遇到一个小问题： 有些任务几分钟就结束，有些却要跑很久。一直守在电脑前没必要，但去做别的事情，又总忍不住隔一会儿回来看看： “跑完了吗？” 既然 Codex 在任务完成或出错时都会发送 Windows 系统通知，能不能把这些通知直接推送到手机上？ WinToastRelay 是一款开源的 Windows...
I Built a Mobile Terminal Around My herdr + Codex Workflow
I Built a Mobile Terminal Around My herdr + Codex Workflow I am building Termish, an open-source mobile remote-work tool for SSH/Mosh terminals, file management, remote screens, and AI-assisted development. The unusual part is that I built much of Termish through Termish itself. My daily workflow is: The development environment and AI tools run on my own machine. My phone is where I enter commands, describe tasks, upload screenshots, inspect changes, and check the resulting UI. Termish is both the product I am building and a tool I use to build it. Put remote work in your pocket. Source code: github.com/ttermish/termish Why do AI-assisted coding on a phone? A phone is not a replacement for a desktop machine. Its screen is smaller. Reading a large codebase is harder. Long debugging...
Which tools do Claude, Codex and Cursor choose? We measured 17k runs to find out
Article URL: https://armature.tech/blog/which-tools-coding-agents-install Comments URL: https://news.ycombinator.com/item?id=49557206 Points: 73 # Comments: 22
Wiring CAI into OpenClaw, Hermes, Codex, and Cursor: the exact MCP config for each host
Wiring CAI into OpenClaw, Hermes, Codex, and Cursor: the exact MCP config for each host The general install is one command, , and the contract lives at . This post is the host-specific deep-dive: the exact config block for OpenClaw, Hermes, Codex, and Cursor, with the path to the config file, the env var handling, and the restart command. Before you start You need: A CAI account (). The signup is four steps, about 2 minutes. An API key from the account dashboard. The key has a scope. Start with the read scopes and add when you need transfers. The MCP server installed (). The MCP host of your choice installed and running. OpenClaw OpenClaw's MCP config lives at . Add the CAI server to the object: Restart OpenClaw. The CAI tools appear in the agent's tool surface. Verify with . The server...
The ChatGPT/Codex app bundles a full copy of LibreOffice
Article URL: https://simonwillison.net/2026/Sep/1/codex-libreoffice/ Comments URL: https://news.ycombinator.com/item?id=49527396 Points: 229 # Comments: 112
How to Connect Codex to a Custom Responses API Provider and Verify the Route
Changing an API base URL is not enough to prove that Codex is using the provider you intended. Codex is an agentic client. It needs more than a model that can return text. The provider must support the Responses API behavior Codex relies on, the API key must reach the process that launches Codex, the selected model must be available to that key, and the final request must arrive at the expected endpoint. This guide shows a conservative setup for a custom Responses API provider, using XiuRouter as the concrete example. The same verification method applies to other compatible gateways. The configuration has four separate decisions A working Codex provider configuration answers four questions: Which model should Codex request? Which named provider should Codex use? Which base URL should...
Make Codex Prove It: A Three-File Design That Leaves Evidence on Disk
An AI agent telling you "done" is not evidence. When I started delegating work to Codex, I took those reports at face value — until I checked the code and found the change missing, the wrong file edited, or no commit at all. So I stopped trusting language and started making the shell write the facts to disk. Why this design works When you hand a task to Codex, it comes back with "Completed." At first that satisfied me. But when I actually checked the code, the critical change wasn't there, or a different file had been touched, or had never run. The output "I did it" and the fact "it was actually done" are two different things. This is true of Claude Code too. Whether tool results were read correctly, whether errors were swallowed — even with code I wrote myself, running a self-audit right...
[Codex] 分享个一直在用的工具：/goal 的进度条
以插件的形式植入 Codex （一个 mcp 一个 skill ），通过 CDP 把结果显示在 Codex 的 Goal 界面 进度计算方式是模型在开始进度前会写一个 CheckList ，根据任务量大小，每一步会有不同的权重， 完成一步，会乘以权重以后再加到进度里， 计算和实时更新都是程序跑的，模型只管勾选 CheckList 就行 所以不怎么费...