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

blog-post博客文章

Agent Skill

blog-post 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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

2,846

周安装

121

GitHub Stars

33

下载量

997
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-post(博客文章)
来源仓库:https://github.com/mblode/agent-skills
仓库路径:skills/blog-post
安装命令:
npx skills add https://github.com/mblode/agent-skills --skill blog-post
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/mblode/agent-skills --skill blog-post

简介

blog-post 用于查找、检索和筛选相关信息,支持关键词定位。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中需要快速定位候选结果时使用。
  • 通过 GitHub 安装,使用 npx skills add 命令添加指定仓库的 skill。
  • 安装前需确认权限范围、维护状态,以及是否会触发联网或文件读写。
  • blog-post 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Blog Post

Research, draft, and polish publication-ready blog posts.

Workflow

Blog post progress:
- [ ] Step 1: Clarify topic and format
- [ ] Step 2: Research and gather sources
- [ ] Step 3: Extract key insights and outline
- [ ] Step 4: Write the post
- [ ] Step 5: SEO and polish
- [ ] Step 6: Validate quality

Step 1: Clarify topic and format

Determine the post type before writing. Ask the user if unclear:

FormatBest forStructure
ListicleSurprising takeaways from sourcesNumbered insights with analysis
Tutorial / how-toTeaching a process step by stepProblem → steps → result
NarrativePersonal experience or journeyStory arc with tension and resolution
Thought leadershipIndustry commentary or opinionThesis → evidence → implications

Step 2: Research and gather sources

Read files, fetch URLs, or accept pasted text. If the user provides a topic without sources, research the topic before writing. Read every source completely before outlining.

Accept messy input — scattered thoughts, bullet points, brain dumps, random observations. The mess is the input; structure is the output.

Step 3: Extract key insights and outline

  • Identify the most surprising, counter-intuitive, or impactful takeaways.
  • Rank insights by reader value — lead with the strongest.
  • Select 2-4 powerful quotes for blockquotes (only the best).
  • Find a relatable problem or curiosity gap to anchor the introduction.
  • Check for narrative potential: is there a journey from one understanding to another?

Create a section outline before drafting. Each section should have a clear purpose.

Step 4: Write the post

Load the matching template from references/format-templates.md and follow its structure. Only load the template for the chosen format — not all four.

Step 5: SEO and polish

  • Title: under 60 characters, include the main keyword.
  • Meta description: 150-160 characters summarizing the post's value.
  • Keyword placement: naturally in the title, first paragraph, and 2-4 times throughout.
  • Headers: descriptive H2/H3 tags that work as standalone scan points.
  • Internal links: suggest 2-3 related links if the user has other content.

Step 6: Validate quality

Before finalizing, verify:
- [ ] Hook grabs attention in the first 1-2 sentences
- [ ] Headline is specific and delivers on its promise
- [ ] Every paragraph is 2-4 sentences maximum
- [ ] Each section has analysis or insight, not just summary
- [ ] Blockquotes are limited to 2-4 total
- [ ] Conclusion looks forward with a question, CTA, or provocative takeaway
- [ ] Tone is conversational throughout — no academic language
- [ ] Title is under 60 characters with the main keyword
- [ ] Post includes a meta description (150-160 chars)
- [ ] Markdown renders correctly (headings, blockquotes, bold, code)

Blog-specific voice

  • Write to a smart friend, not a committee. Authenticity beats authority.
  • Bold key phrases for scannability — readers skim before they read.
  • Show, don't tell: specific examples with real details beat vague claims.
  • Admit uncertainty when genuine. Never bluff expertise.

Headline patterns

  • Number + insight: "7 Things [Sources] Reveal About [Topic]"
  • Surprising tension: "Why [Common Belief] Gets [Topic] Wrong"
  • How-to framing: "How to [Achieve Outcome] with [Method]"
  • Question format: "What If [Provocative Reframe]?"

Under 60 characters for SEO. Sentence case, no trailing period. Must deliver on its promise.

Anti-patterns

  • Generic headlines ("Interesting Findings About X").
  • Conclusions that just summarize what was already said.
  • Clickbait headlines the content does not deliver on.
  • Blockquoting every source instead of choosing the 2-4 most powerful.
  • Skipping research and writing from assumptions.

Skill handoffs

WhenRun
After blog post is written, audit prose qualitydocs-writing
If post needs to become a presentationpresentation-creator
Edit and polish the copycopy-editing
Optimize SEO beyond basicsoptimise-seo

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

Claude

33.94%
按下载量换算338

Codex

33.38%
按下载量换算333

Cursor

20.79%
按下载量换算207

Gemini CLI

9.15%
按下载量换算91

安全审计

Gen Agent Trust Hub

通过

Socket

通过

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