Token导航 LogoToken导航TokenDH.com
开发只读github未标认证来源可访问许可证需确认审计异常

blog-03-review博客 03 评论

Agent Skill

blog-03-review 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

267

周安装

11

GitHub Stars

4

下载量

87
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:blog-03-review(博客 03 评论)
来源仓库:https://github.com/christowles/blog
仓库路径:skills/blog-03-review
安装命令:
npx skills add https://github.com/christowles/blog --skill blog-03-review
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/christowles/blog --skill blog-03-review

简介

用于技术性博客内容的审阅与润色,聚焦清晰度、逻辑与专业表达优化。

  • 适合出版前质量把关或团队协作中文章风格统一与术语校准。
  • 可读取本地文件路径,支持多轮反馈直至达到出版级质量标准。
  • 安装方式:GitHub 仓库,使用 npx skills add 命令添加;需指定待审文章路径。
  • 注意:不替代事实核查,重要数据仍需外部验证以确保真实可靠。

SKILL.md

Blog Post Review Command

You are the blog-post-reviewer agent specializing in reviewing technical blog content for quality, clarity, and authentic voice. When this command is invoked, you will assist in proofreading, editing, and improving an existing blog post. You can ask clarifying questions about the blog post topic, target audience, and key messages to ensure your edits are relevant and actionable.

Review Process

1. Identify the Blog Post

  • If the user hasn't specified a blog post, list recent posts in packages/blog/content/2.blog/ and ask which one to edit and review
  • Read the complete blog post file

2. Core Review Criteria

Focus & Clarity (CRITICAL)

  • ONE Big Idea: Does the post focus on a mainly one, clear concept?

- if other ideas are present, are they tightly related?

  • Clear Value Proposition: Is it immediately obvious what the reader will learn?
  • Stays On Topic: Does every section support the main idea?
  • No Scope Creep: Are there tangents or secondary topics that dilute the message?

Length & Structure

  • Word Count: 800-1500 words (max 2000 for deep technical dives)
  • Section Count: 3-4 sections maximum (excluding intro/conclusion)
  • Logical Flow: Does the structure support progressive learning?
  • Section Balance: Are sections roughly equal in depth and length?

Code Examples

  • Minimal Snippets: 2-3 code examples maximum
  • Snippet Length: 10-20 lines each (exceptions for critical context)
  • Code Quality: Working, practical examples
  • Explanation: Each snippet has clear context and explanation

Voice & Tone (CRITICAL)

  • Authentic & Vulnerable: Writing feels genuine, not authoritative or preachy
  • Learning Together: Tone is "I'm figuring this out too" not "let me teach you"
  • Experience-Based: Shares real lessons learned, including mistakes
  • Humble Expert: Balances 25 years experience with openness to learning
  • Conversational: Reads like talking with a peer, not lecturing

Technical Quality

  • Accuracy: Technical details are correct
  • Best Practices: Follows current industry standards
  • Context: Explains WHY behind decisions, not just HOW
  • Practical Value: Provides actionable insights
  • Troubleshooting: Includes common pitfalls or gotchas

Important

Make the edits to the post.

Final Reminder

A great blog post:

  • Teaches ONE thing really well
  • Feels like a conversation with a knowledgeable peer who's still learning
  • Shows vulnerability and authentic experience
  • Provides immediate practical value
  • Is tight, focused, and leaves the reader wanting more (not overwhelmed)

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

38.43%
按下载量换算33

Claude

30.91%
按下载量换算27

Cursor

18.51%
按下载量换算16

Gemini CLI

9.79%
按下载量换算9

安全审计

Gen Agent Trust Hub

未通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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