Token导航 LogoToken导航TokenDH.com
开发需要联网github未标认证来源可访问许可证需确认审计提醒

codexCodex 编程助手

Agent Skill

codex 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

218

周安装

9

GitHub Stars

公开资料未说明

下载量

71
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:codex(Codex 编程助手)
来源仓库:https://github.com/abpai/skills
仓库路径:skills/codex
安装命令:
npx skills add https://github.com/abpai/skills --skill codex
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/abpai/skills --skill codex

简介

codex 是 Codex 编程助手的核心技能,支持任务执行、代码生成与沙盒环境交互。

  • 适用于新任务启动或继续中断会话,提供模型选择、推理强度与读写权限控制。
  • 通过 codex exec 命令运行,默认启用 --skip-git-repo-check 并抑制 stderr 输出以稳定输出流。
  • 涉及文件操作与网络请求时需明确指定 sandbox 模式,注意数据安全与资源限制。
  • codex 属于开发类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Codex Skill Guide

Workflow

  1. Confirm task mode:

- New run: use codex exec. - Continue prior run: use codex exec... resume --last with stdin prompt.

  1. Set defaults unless user overrides:

- Model: gpt-5.4. - Reasoning effort: ask user to choose xhigh, high, medium, or low. - Sandbox: read-only unless edits/network are required.

  1. Build command with required flags:

- Always include --skip-git-repo-check. - Add 2>/dev/null by default to suppress thinking tokens on stderr. - Show stderr only if user asks or debugging is needed.

  1. Run command, summarize outcome, and ask what to do next.
  2. After completion, remind user they can continue with codex resume.

Quick Reference

Use caseSandbox modeKey flags
Read-only review or analysisread-only--sandbox read-only 2>/dev/null
Apply local editsworkspace-write--sandbox workspace-write --full-auto 2>/dev/null
Permit network or broad accessdanger-full-access--sandbox danger-full-access --full-auto 2>/dev/null
Resume recent sessionInherited from original`echo "prompt" \codex exec --skip-git-repo-check resume --last 2>/dev/null`
Run from another directoryMatch task needs-C <DIR> plus other flags 2>/dev/null

Command Patterns

New run

codex exec --skip-git-repo-check \
  --model gpt-5.4 \
  --config model_reasoning_effort="high" \
  --sandbox read-only \
  "your prompt here" 2>/dev/null

Resume latest session

Use stdin and keep flags between exec and resume.

echo "your prompt here" | codex exec --skip-git-repo-check resume --last 2>/dev/null

When resuming, do not add configuration flags unless the user explicitly asks for changes (for example, different model or reasoning effort).

Model Options

ModelBest forContext windowKey features
gpt-5.4Default for most coding tasks in CodexN/A in this skillOpenAI's recommended default for general-purpose coding
gpt-5.4-proHarder problems that benefit from more computeN/A in this skillMore compute for deeper reasoning on difficult tasks
gpt-5-miniFaster/cost-effective option for lighter tasksN/A in this skillSmaller GPT-5 model for lower-cost coding and chat tasks
gpt-5.3-codexLegacy specialized alternativeN/A in this skillPrior Codex-tuned model; generally superseded by GPT-5.4

gpt-5.4 is the default for software engineering tasks.

Reasoning Effort

  • xhigh - Ultra-complex tasks (deep problem analysis, complex reasoning, deep understanding of the problem)
  • high - Complex tasks (refactoring, architecture, security analysis, performance optimization)
  • medium - Standard tasks (refactoring, code organization, feature additions, bug fixes)
  • low - Simple tasks (quick fixes, simple changes, code formatting, documentation)

Following Up

  • After every run, ask for next steps or clarifications.
  • When proposing another run, restate model, reasoning effort, and sandbox mode.
  • For continuation, use stdin with resume --last.

Error Handling

  • If codex --version or codex exec exits non-zero, report failure and ask before retrying.
  • Ask permission before high-impact flags unless already granted: --full-auto, --sandbox danger-full-access, --skip-git-repo-check.
  • If output includes warnings or partial results, summarize and ask how to proceed.

CLI Version

Use a current Codex CLI version that supports gpt-5.4. Check with:

codex --version

Use /model inside Codex to switch models, or set defaults in ~/.codex/config.toml.

Update Check

On first use in a session, silently check for a newer version:

  1. Fetch https://raw.githubusercontent.com/abpai/skills/main/versions.json.
  2. Compare the version for codex against this file's metadata.version.
  3. If the remote version is newer, pause before the main task and ask: codex update available (local {X.Y} → remote {A.B}). Would you like me to update it for you first? I can run npx skills update codex for you.
  4. If the user says yes, run the update before continuing.
  5. If the user says no, continue with the current local version.
  6. If the fetch fails or web access is unavailable, skip silently.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

36.06%
按下载量换算26

Claude

29.02%
按下载量换算21

Cursor

19.46%
按下载量换算14

Gemini CLI

10.38%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

可疑

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills