[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"pack-detail-codex-cli-quick-start-en":3,"seo:pack:codex-cli-quick-start:en":103},{"code":4,"message":5,"data":6},200,"操作成功",{"pack":7},{"slug":8,"icon":9,"tone":10,"status":11,"status_label":12,"title":13,"description":14,"items":15,"install_cmd":102},"codex-cli-quick-start","⚡","#10A37F","new","New · this week","Codex CLI Quick Start","Ten picks for the developer who just installed OpenAI Codex CLI and wants to ship in the first hour. Codex itself, AGENTS.md, sandbox runtime, MCP servers, the plugins\u002Fskills marketplaces, a zero-config helper, and the sister-product Claude Code bridge — in install order.",[16,28,38,47,57,65,73,81,88,95],{"id":17,"uuid":18,"slug":19,"title":20,"description":21,"author_name":22,"view_count":23,"vote_count":24,"lang_type":25,"type":26,"type_label":27},639,"11647c80-61e5-40ba-951e-69f433cabc1c","openai-codex-cli-terminal-coding-agent-openai-11647c80","OpenAI Codex CLI — Terminal Coding Agent by OpenAI","Official OpenAI terminal coding agent. Reads your codebase, writes code, runs commands, and manages files from the command line. Multi-model support with sandbox execution. 20,000+ stars.","OpenAI",264,0,"en","script","Script",{"id":29,"uuid":30,"slug":31,"title":32,"description":33,"author_name":34,"view_count":35,"vote_count":24,"lang_type":25,"type":36,"type_label":37},3134,"ddb121a1-a9a0-4151-920e-e787fc206927","agents-md-open-format-for-coding-agent-instructions","AGENTS.md — Open Format for Coding Agent Instructions","AGENTS.md defines a predictable instruction file for coding agents: dev env tips, tests, PR rules, and repo context. Drop it at repo root.","Prompt Lab",76,"skill","Skill",{"id":39,"uuid":40,"slug":41,"title":42,"description":43,"author_name":22,"view_count":44,"vote_count":24,"lang_type":25,"type":45,"type_label":46},96,"26b9b7dd-dbe6-41e3-a093-17db5409d739","openai-cookbook-official-prompting-guides-26b9b7dd","OpenAI Cookbook — Official Prompting Guides","Official prompting guides from OpenAI: GPT-5.2, Codex, Meta Prompting, and Realtime API guides. The definitive reference for OpenAI model optimization.",255,"prompt","Prompt",{"id":48,"uuid":49,"slug":50,"title":51,"description":52,"author_name":53,"view_count":54,"vote_count":24,"lang_type":25,"type":55,"type_label":56},42,"8a28360d-6eef-449d-9857-d8b388a05f34","mcp-reference-servers-official-collection-8a28360d","MCP Reference Servers — Official Collection","Official MCP reference server implementations by Anthropic: filesystem, git, memory, fetch, sequential thinking, and more. The starting point for MCP development.","MCP Hub",127,"mcp","MCP",{"id":58,"uuid":59,"slug":60,"title":61,"description":62,"author_name":63,"view_count":64,"vote_count":24,"lang_type":25,"type":55,"type_label":56},3690,"4c7e217d-bc3a-5018-823c-3aef3386f9cb","aio-sandbox-secure-agent-runtime-in-docker","AIO Sandbox — Secure Agent Runtime in Docker","AIO Sandbox ships a browser\u002Fshell\u002Ffiles\u002FMCP runtime in one container; verified 4,666★ and offers a 30-second Docker quick start on port 8080.","AI Open Source",75,{"id":66,"uuid":67,"slug":68,"title":69,"description":70,"author_name":71,"view_count":72,"vote_count":24,"lang_type":25,"type":36,"type_label":37},3683,"73fa822d-b601-5594-9905-0caeab93094c","awesome-codex-plugins-curated-plugin-marketplace","Awesome Codex Plugins — Curated Plugin Marketplace","Awesome Codex Plugins curates installable bundles for Codex and peers; verified 51 stars with marketplace JSON, lint commands, and mirrored plugin paths.","Skill Factory",55,{"id":74,"uuid":75,"slug":76,"title":77,"description":78,"author_name":79,"view_count":80,"vote_count":24,"lang_type":25,"type":26,"type_label":27},3267,"1f31559c-5f92-4931-b31c-099a72420b67","zcf-zero-config-setup-for-claude-code-codex","zcf — Zero-Config Setup for Claude Code & Codex","Zero-config initializer for Claude Code and Codex: one `npx zcf i` sets up workflows, routing, and optional MCP pieces in minutes.","Agent Toolkit",70,{"id":82,"uuid":83,"slug":84,"title":85,"description":86,"author_name":71,"view_count":87,"vote_count":24,"lang_type":25,"type":36,"type_label":37},3682,"823cacb7-fa6b-5a0c-842d-73dd6a9a45f1","awesome-qa-skills-test-skills-for-codex-claude","Awesome QA Skills — Test Skills for Codex & Claude","Awesome QA Skills organizes reusable testing skills for Codex, Cursor, and Claude Code; verified 58 stars with copy-based installs and all-tool scripts.",234,{"id":89,"uuid":90,"slug":91,"title":92,"description":93,"author_name":79,"view_count":94,"vote_count":24,"lang_type":25,"type":36,"type_label":37},788,"208207e5-4645-4bc4-b8d6-c5e3be2e9803","codex-plugin-claude-code-openai-cross-review-208207e5","Codex Plugin for Claude Code — OpenAI Cross-Review","Official OpenAI plugin that adds Codex code review and task delegation inside Claude Code. Get adversarial reviews, rescue debugging, and multi-model collaboration.",186,{"id":96,"uuid":97,"slug":98,"title":99,"description":100,"author_name":34,"view_count":101,"vote_count":24,"lang_type":25,"type":45,"type_label":46},804,"a4647a34-1fb8-406e-8809-ef1509e08b4b","cursor-vs-claude-code-vs-codex-ai-coding-compared-a4647a34","Cursor vs Claude Code vs Codex — AI Coding Compared","In-depth comparison of the three leading AI coding tools in 2026. Covers pricing, context window, MCP support, agent capabilities, and best use cases for each platform.",313,"tokrepo install pack\u002Fcodex-cli-quick-start",{"pageType":104,"pageKey":8,"locale":25,"title":105,"metaDescription":106,"h1":107,"tldr":108,"bodyMarkdown":109,"faq":110,"schema":126,"internalLinks":131,"citations":144,"wordCount":157,"generatedAt":158},"pack","Codex CLI Quick Start — 10 Picks to Ship in the First Hour","Just installed OpenAI Codex CLI? Ten opinionated picks — Codex itself, AGENTS.md, OpenAI Cookbook, MCP reference servers, AIO Sandbox, the Codex plugin marketplace, zcf zero-config, QA skills, the Claude Code cross-review plugin, and a tool-comparison primer. Install in this order and ship something useful before lunch.","Codex CLI Quick Start — A First-Hour Setup for OpenAI's Coding Agent","Ten picks in deliberate order: install Codex, drop an AGENTS.md, bookmark the OpenAI Cookbook, wire MCP reference servers, decide your sandbox story, browse the plugin marketplace, run zcf for zero-config, layer in QA skills, install the Claude Code cross-review plugin, and read the Cursor\u002FClaude Code\u002FCodex comparison so you know when to reach for which. By the end you have a Codex CLI that knows your repo, runs in a real sandbox, talks to MCP servers, and pairs with Claude Code when you want a second opinion.","## What's in this pack\n\nYou just ran `npm install -g @openai\u002Fcodex`, typed `codex`, and now you're staring at a Rust-fast prompt wondering what to do next. Codex CLI feels obviously powerful out of the box — but the first hour decides whether you ship something real or close the tab. This pack is the **opinionated short list** that fixes the first ten problems every new Codex user actually hits.\n\nCodex is the sister product to Claude Code: both are terminal coding agents, both speak the [Agent Skills](https:\u002F\u002Fdocs.anthropic.com\u002Fen\u002Fdocs\u002Fagents-and-tools\u002Fagent-skills) standard, both read a per-repo instruction file. Codex's differentiator is its **strong CLI ergonomics**, its **multi-model support**, and its **opt-in sandbox** for running commands. This pack leans into that: project memory first (AGENTS.md), then the runtime defaults you want to lock down (sandbox), then the marketplaces and helpers, then the bridge back to Claude Code for cross-review.\n\nWho this is for: a developer with a working terminal, a real codebase, an OpenAI API key (or a Codex subscription), and zero patience for browsing every plugin in alphabetical order. By the time you finish step 10, you'll have a Codex install that knows your repo conventions, runs in a real sandbox, talks to MCP servers, and pairs with Claude Code when you want a second opinion on a tricky diff.\n\n## Install in this order\n\n1. **OpenAI Codex CLI — the agent itself** — Start here. `npm install -g @openai\u002Fcodex` then `codex` in any repo. Rust-based, supports the Agent Skills standard, multi-model (GPT-5, o-series, custom), and ships with sandboxed execution. Get a session running before you install anything else — every later pick configures *this* tool.\n2. **AGENTS.md — the project memory file** — The single highest-leverage file you'll ever write for Codex. Drop a short markdown at repo root: dev-env tips, test commands, PR conventions, things the agent **must not touch**. Codex reads it on every session. Same idea as CLAUDE.md, but cross-tool: Cursor, Continue, and other agents read AGENTS.md too. **5 minutes to fill in. Saves hours every week.**\n3. **OpenAI Cookbook — Official Prompting Guides** — Codex's behavior is heavily shaped by how you ask. The Cookbook has the official prompting guides for GPT-5, Codex, meta-prompting, and the Realtime API. Skim the Codex-specific section once before you start cargo-culting prompt patterns from random tweets.\n4. **MCP Reference Servers — Official Anthropic Collection** — Codex speaks MCP. Connect the filesystem, git, memory, fetch, and sequential-thinking reference servers and you've unlocked the same MCP ecosystem Claude Code uses. These are *reference* implementations — small, auditable, good starting points before you reach for third-party MCP servers.\n5. **AIO Sandbox — Secure Agent Runtime in Docker** — Decide your sandbox story early. Codex's built-in sandbox is good for *file* writes; AIO Sandbox containerizes the *whole* runtime (browser + shell + files + MCP) so a misbehaving session can't touch your laptop. 30-second Docker quick-start on port 8080. Install it before you let Codex auto-run shell commands on real code.\n6. **Awesome Codex Plugins — Curated Marketplace** — A community-curated list of installable plugin bundles. This is your browse-instead-of-grep index for the next six months. Skim once, bookmark the categories that match your stack, come back when you need something specific. (Codex calls these *plugins*; Claude Code calls them *skills* — same shape, slightly different packaging.)\n7. **zcf — Zero-Config Setup for Claude Code & Codex** — `npx zcf i` and it sets up workflows, routing rules, and MCP wiring in minutes. Useful if step 2-5 felt like a lot of manual config. It opinionates so you don't have to think about each toggle. Best on a fresh setup; don't run it on top of careful hand-tuning.\n8. **Awesome QA Skills — Test Skills for Codex & Claude** — A reusable testing-skills bundle that works across Codex, Cursor, and Claude Code. Once Codex starts writing code, you want a test pass that runs the same way every time. Drop these in and Codex stops shipping subtle regressions.\n9. **Codex Plugin for Claude Code — OpenAI Cross-Review** — The bridge. Install this inside Claude Code and you can hand a diff to Codex for an adversarial review, or delegate a sub-task. Two coding agents on the same problem catch different bugs. Pairs especially well for tricky refactors where one model gets stuck.\n10. **Cursor vs Claude Code vs Codex — 2026 Comparison** — Read this last. After you've installed Codex and the bridge to Claude Code, this primer tells you which agent to reach for in which situation: context window, MCP support, pricing, sandbox model, agent ergonomics. It's not about picking a winner — it's about knowing the tradeoffs you just opted into.\n\n## How they fit together\n\n```\nCodex CLI (#1)\n   │\n   └─ the agent, reads everything below\n         │\nAGENTS.md (#2)\n   │\n   └─ per-repo memory, loaded every session\n         │\nOpenAI Cookbook (#3)\n   │\n   └─ how to write prompts Codex actually follows\n         │\nMCP Reference Servers (#4)\n   │\n   ├─ filesystem, git, memory, fetch\n   └─ structured access to your environment\n         │\nAIO Sandbox (#5)\n   │\n   └─ containerized runtime so shell commands\n      can't escape into your dotfiles\n         │\nMarketplaces: Awesome Codex Plugins (#6)\n   │\n   └─ browse instead of guess\n         │\nzcf (#7)  — optional zero-config helper\n   │\n   └─ wires #2\u002F#4\u002F#6 if you skipped manual setup\n         │\nQA Skills (#8)\n   │\n   └─ post-edit testing so Codex stops shipping regressions\n         │\nCodex Plugin in Claude Code (#9)\n   │\n   └─ cross-review: Codex critiques Claude's diff or vice versa\n         │\nCursor\u002FClaude Code\u002FCodex comparison (#10)\n   │\n   └─ when to reach for which agent\n```\n\nThe **Codex + AGENTS.md + AIO Sandbox** trio is the load-bearing wall. Everything else is decoration without those three. If you only have 10 minutes, install #1, write a 20-line #2, and decide on #5 before you let Codex run a single shell command. The marketplace browsing (#6, #8) and the cross-product bridge (#9) are upgrades for week two.\n\n## Tradeoffs you'll hit\n\n- **Codex CLI vs Claude Code** — Both are terminal coding agents, both speak Agent Skills, both read a per-repo instruction file. Codex leans on OpenAI models (GPT-5, o-series) and has the cleaner sandbox story today; Claude Code leans on Anthropic models and has the deeper subagent + hooks ecosystem. Most serious users install both — Codex for raw speed and OpenAI-specific tasks, Claude Code for orchestrated long sessions — and use the cross-review plugin (#9) to play them against each other.\n- **AGENTS.md vs CLAUDE.md** — Same idea, different conventions. AGENTS.md is the cross-tool open format (Codex, Cursor, Continue, others read it). CLAUDE.md is Claude Code's own file. If you only run one agent, use that agent's native file. If you run both, write AGENTS.md and have CLAUDE.md `@import` it — single source of truth.\n- **Codex built-in sandbox vs AIO Sandbox (#5)** — Codex's built-in sandbox protects *file writes* and gates network access; it's process-level. AIO Sandbox wraps the *entire runtime* in a container — browser, shell, files, MCP. Built-in is enough for solo tinkering on your own code. AIO is what you want before you let an agent run arbitrary shell on a real machine — especially on a work laptop or production-adjacent host.\n- **Reference MCP servers (#4) vs third-party MCP** — Reference servers are Anthropic-authored, small, auditable, and have boring scopes (filesystem, git, fetch). Third-party MCP servers are often more capable but vary wildly in code quality. Always start with reference servers — they're the floor that everything else gets compared against.\n- **zcf (#7) vs hand-tuning** — zcf gets you 80% of a sensible setup in 60 seconds. The other 20% is the part that matters to *your* repo: which MCP servers you want, what AGENTS.md should actually say, which sandbox model fits your security posture. Run zcf for the scaffolding, then hand-edit the parts that survive a project review.\n\n## Common pitfalls\n\n- **Skipping AGENTS.md because Codex \"seems smart enough\"** — Without it, every session re-learns your conventions. The agent will keep using tabs in a spaces repo, run `pytest` in a project that uses `uv run pytest`, and edit files you told three previous sessions not to touch. 30 lines of markdown solves it permanently.\n- **Running Codex outside a sandbox on a work machine** — The default is reasonable, but \"reasonable\" is not \"safe enough for a misconfigured prompt at 2am.\" AIO Sandbox (#5) takes 30 seconds and removes a whole class of bad days.\n- **Treating AGENTS.md as a long doc** — 30–80 lines is the sweet spot. People paste their architecture doc in and burn thousands of tokens every turn. Keep it terse: stack, commands, conventions, hard NOs. Move long-form architecture into linked docs that the agent can read on demand.\n- **Installing every plugin from the marketplace (#6) at once** — More plugins = more context Codex has to scan to route correctly = slower and dumber routing. Start with the four or five that match your stack. Add only when you hit a real workflow gap.\n- **Ignoring the cross-review plugin (#9) until you're stuck** — The right time to install it is before you need it. The wrong time is at hour three of a debugging session when Codex and Claude Code are both spinning on the same wrong hypothesis and you have to context-switch to install something just to break the loop.",[111,114,117,120,123],{"q":112,"a":113},"Do I really need 10 things just to start using Codex CLI?","No — Codex works fine straight out of `npm install -g @openai\u002Fcodex`. This pack is for the moment after that, when you realize you're repeating yourself every session, you can't decide whether to use the built-in sandbox or AIO Sandbox, and you want the boring infrastructure layer set up properly once. If you're still in the \"try the prompt for an hour\" phase, just use Codex raw. Come back when you've hit your third \"I wish I'd configured this earlier.\"",{"q":115,"a":116},"Which three would you install if I only have 10 minutes?","Codex itself (#1), a 20-line AGENTS.md (#2), and a decision on sandbox (#5 — at minimum, know whether you're going to use the built-in or run inside AIO Sandbox). Those three are the minimum viable setup that punches above its weight. The other seven are upgrades you can layer in over the next week.",{"q":118,"a":119},"Codex CLI vs Claude Code — should I pick one?","Honest answer: install both. They overlap a lot (both terminal coding agents, both speak Agent Skills, both read a per-repo instruction file) but they're not interchangeable. Codex is faster on raw token throughput today, has the cleaner sandbox story, and gives you OpenAI models. Claude Code has the deeper subagent + hooks ecosystem and the longer effective context. The cross-review plugin (#9) lets them critique each other's diffs — which is genuinely useful, not just a novelty.",{"q":121,"a":122},"What's the difference between AGENTS.md and CLAUDE.md?","Same idea, different conventions. AGENTS.md is an open cross-tool format — Codex, Cursor, Continue, and others all read it. CLAUDE.md is Claude Code's own file. If you only use one agent, use that agent's native file. If you use both, write AGENTS.md as your source of truth and have CLAUDE.md `@import` it. Don't maintain two copies.",{"q":124,"a":125},"Is the built-in Codex sandbox enough, or do I really need AIO Sandbox?","Built-in is enough for solo tinkering on your own code on a personal machine — it gates file writes and network access at the process level. AIO Sandbox wraps the whole runtime (browser, shell, files, MCP) in a container so even a worst-case prompt can't reach your dotfiles, your SSH keys, or your cloud credentials. For a work laptop or anything production-adjacent, the 30-second Docker setup is worth it. For weekend hacking on a side project, built-in is fine.",{"@context":127,"@type":128,"name":13,"description":129,"numberOfItems":130,"inLanguage":25},"https:\u002F\u002Fschema.org","ItemList","Ten opinionated picks a brand-new OpenAI Codex CLI user should install in their first hour, in install order.",10,[132,136,140],{"url":133,"anchor":134,"reason":135},"\u002Fen\u002Ftopics\u002Fclaude-code-onboarding-30min","Claude Code 30-Minute Onboarding","Sister-product pack for users who also run Claude Code",{"url":137,"anchor":138,"reason":139},"\u002Fen\u002Fai-tools-for\u002Fcoding","AI tools for coding","Compare Codex CLI with Cursor, Claude Code, and other coding agents",{"url":141,"anchor":142,"reason":143},"\u002Fen\u002Ffeatured","Featured assets on TokRepo","After the starter pack, browse the broader curated catalog",[145,149,153],{"claim":146,"source_name":147,"source_url":148},"Codex CLI is an open-source OpenAI terminal coding agent supporting Agent Skills, multi-model, and sandboxed execution","openai\u002Fcodex on GitHub","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fcodex",{"claim":150,"source_name":151,"source_url":152},"AGENTS.md is an open instruction-file format adopted across multiple coding agents","AGENTS.md format site","https:\u002F\u002Fagents.md\u002F",{"claim":154,"source_name":155,"source_url":156},"MCP reference servers (filesystem, git, memory, fetch) are maintained as the canonical implementations of the Model Context Protocol","modelcontextprotocol\u002Fservers on GitHub","https:\u002F\u002Fgithub.com\u002Fmodelcontextprotocol\u002Fservers",1551,"2026-05-22T00:00:00Z"]