#Codex

Language

5 posts

AI & Data•Aug 03, 2026•10 min readZH·EN

我去验证「Codex 不会自省」，最后发现是我自己的程序先挂了电话

一篇博文说 Codex 的主 session 没有纠错的自省能力，Claude 才有，还建议 context 别超过 300k。我刚把 Kimi K3 接进自己的 Web Claude Code Pilot，正想知道它能不能当 orchestrator，就去验了一遍。k3 和 Sol 都有自省能力；那条 300k 的建议也是空的，因为 Codex 在 258.4k 就自己压缩了。而「Codex 主 session 没产出」的症状，是我自己的客户端提前关流造成的——那个 bug 只长在 Codex 那条通道上。

AI AgentsCodexGPT-5.6Kimi K3Claudesubagentorchestrator上下文窗口Web Claude Code Pilot

Programming•Apr 25, 2026•10 min readEN

Three Tools, One Repo: Governing AI Coding When Your Team Won't Standardize

Your teammates use Claude Code, Copilot, and Codex. The enterprise plans don't fix the problem. Here's a four-layer model for governing AI coding at the project and task level — and why the repo, not the tool, has to be the source of truth.

AI CodingTeam WorkflowAGENTS.mdCLAUDE.mdClaude CodeGitHub CopilotCodexVibe CodingGovernancePre-commit HooksMonorepo

Build•Mar 30, 2026•9 min readZH

Claude 开始省着用以后，我把 Codex 捡回来了

一开始我对 Codex 的印象并不好。用惯了 Claude Code 之后，更觉得它差得远。但在 Claude Code 的优惠结束、session 限制变紧之后，我抱着备胎心态重新试了试 Codex，结果 GPT-5.4 反而刷新了我的判断，也直接推动我把它接进了自己的 Web Claude Code Pilot。

AICodexGPT-5.4Claude CodeCodePilot多模型切换SQLiteSide Project

Build•Feb 18, 2026•18 min readEN

Agent Skills: What's Open-Sourced, What's Not, and How to Build Your Own Runtime

The Agent Skills spec is open. The example skills are open. But what about the runtime that loads, selects, and orchestrates them? Every major tool — Claude Code, Codex, Copilot, OpenClaw — built their own. Here's what's open, what's not, and what you actually need to build.

AI AgentsSkillsOpen SourceMCPAnthropicClaude CodeAgent SDKOpenClawCodexCopilotImplementationArchitecture