Token导航 LogoToken导航

Agent Skill · 研究检索

content-marketing(内容营销)

用于辅助文档、README、Markdown、说明文和内容稿件的整理与改写。它适合让 Agent 提炼结构、补齐章节、统一术语、检查链接或把零散材料整理成可读文档。使用时应保留项目已有事实、命令和路径,不要把未确认的信息写成确定结论;涉及对外文案时,还需要控制语气,避免过度营销或夸大能力。

content-marketing 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 48,042。

总安装 706周安装 30GitHub Stars 3下载 247核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

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:content-marketing(内容营销)
来源仓库:https://github.com/oldwinter/skills
仓库路径:skills/content-marketing
安装命令:
npx skills add https://github.com/oldwinter/skills --skill content-marketing
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/oldwinter/skills --skill content-marketing

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

content-marketing 用于辅助文档、README 和内容稿件的整理与改写。适合提炼结构、补齐章节、统一术语或检查链接,提升可读性。通过 npx skills add 命令从指定仓库安装并使用。使用时应保留项目事实,避免夸大能力或写成未确认的结论。

适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/content-marketing

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/oldwinter/skills --skill content-marketing;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

content-marketing 怎么安装?

在终端执行 npx skills add https://github.com/oldwinter/skills --skill content-marketing。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

content-marketing 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

content-marketing 需要什么权限?

权限级别是需要联网。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。

content-marketing 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

content-marketing 用的人多吗?

公开数据是累计安装 706 次,近一周安装 30 次,下载 247 次,GitHub 3 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

content-marketing 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
content-marketing

内容营销 · 本页

7063需要联网未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Content Marketing

Scope

Covers

  • Content market fit (treat content like a product; audience anxieties + needs)
  • SEO-led content (validate search demand before writing)
  • Founder/executive-led thought leadership with a human voice
  • Blog-as-press-release announcements (shareable, SEO-friendly)
  • AI-assisted content workflows with human-in-the-loop quality controls
  • Backlog, editorial calendar, content briefs, distribution, and measurement

When to use

  • “Create a content strategy / content marketing plan.”
  • “Build an SEO plan + editorial calendar.”
  • “We need founder-led thought leadership; pick a channel + voice.”
  • “Turn an announcement into a blog post instead of a press release.”
  • “Create content briefs and a repeatable content production system.”

When NOT to use

  • You still need to define positioning/ICP (use positioning-messaging or problem-definition).
  • You need a technical SEO audit (crawl/indexing, performance, schema, internal linking) more than a content program.
  • You need a paid acquisition strategy (ads, bidding, creative testing) rather than owned content.
  • You cannot publish without a review process and cannot provide one (this skill requires a compliant workflow).

Inputs

Minimum required

  • Product: what it is + who it’s for (ICP/audience)
  • Goals + timebox (e.g., pipeline, awareness, signups, recruiting) and 1 primary metric
  • Primary offer(s) and CTA(s) (demo, trial, newsletter, download)
  • Constraints: team capacity, SME availability, brand/compliance guardrails, regions/languages
  • Spokesperson options (founder/executive/PM/other) and preferred channels (blog, LinkedIn, YouTube, podcast, newsletter)

Missing-info strategy

  • Ask up to 5 questions from references/INTAKE.md, then proceed with explicit assumptions.
  • If SEO demand data is unavailable, produce keyword/topic hypotheses with confidence labels and a demand-validation to-do list (no credentials required).
  • Never request secrets or credentials; accept redacted exports/screenshots if offered.

Outputs (deliverables)

Produce a Content Marketing Plan Pack (Markdown in-chat; or as files if requested) containing:

  1. Context snapshot (ICP, goal, metric, timebox, constraints)
  2. Content market fit brief (audience anxieties, jobs-to-be-done, “why now”)
  3. Channel + voice strategy (human spokesperson, primary channel focus, tone rules, repurposing plan)
  4. SEO demand-validated topic map (topics/keywords, intent, proof of demand, SERP angle)
  5. Backlog + editorial calendar (4–8 weeks, prioritized)
  6. 3 content briefs (ready for writing) + per-piece distribution plan
  7. Announcement blog post template (press-release alternative) + 1 outline/draft (if relevant)
  8. AI-assisted content SOP (AI roles + human review + “information gain” rules)
  9. Measurement plan (leading indicators, dashboards, iteration cadence)
  10. Risks / Open questions / Next steps (always included)

Templates and checklists:

Workflow (8 steps)

1) Intake + success definition

  • Inputs: User prompt; references/INTAKE.md.
  • Actions: Confirm the business goal, ICP, primary CTA, and timebox. Define one primary success metric and 2–3 leading indicators. Confirm constraints (capacity, compliance, languages, distribution).
  • Outputs: Context snapshot (v1).
  • Checks: Goal is measurable and timebound (e.g., “Increase qualified demo requests from SEO by 25% in 8 weeks”).

2) Define content market fit (audience anxieties → promises)

  • Inputs: ICP/audience, pains/goals, buying context, objections.
  • Actions: Treat content as a product: define audience segment(s), top anxieties (career, risk, status, time), and the “promotion narrative” (how content helps them win). Identify 3–5 core “promises” your content should consistently deliver.
  • Outputs: Content market fit brief.
  • Checks: Each promise is specific and maps to a content type (how-to, teardown, case study, POV, template).

3) Validate demand + pick topic themes (SEO + non-SEO)

  • Inputs: Product surfaces, promises, known customer questions, any keyword data.
  • Actions: Build a topic universe, then split into:

- SEO topics: only keep topics with evidence of search demand. - Thought leadership topics: publish for trust/brand even without search demand, but tie to distribution plan and spokesperson voice. Document “evidence of demand” (Search Console, autocomplete, competitor pages ranking, keyword tools, internal tickets).

  • Outputs: Topic map (with demand evidence + confidence labels).
  • Checks: No SEO topic is approved without a demand signal; every non-SEO topic has a distribution owner + channel.

4) Choose the human voice + primary channel

  • Inputs: Spokesperson options, team strengths, audience media habits.
  • Actions: Pick one primary channel that matches the spokesperson’s natural style (long-form writing, short-form posts, video, audio). Define voice rules (first-person, specific opinions, vulnerability/honesty) and what to avoid (corporate ghost tone, generic platitudes). Decide how content will be repurposed across secondary channels.
  • Outputs: Channel + voice strategy (including “say this / not that”).
  • Checks: There is a single primary channel for the next 4–8 weeks; repurposing does not create new work without owners.

5) Build the backlog + editorial calendar

  • Inputs: Topic map, capacity, seasonality, launches/announcements.
  • Actions: Prioritize using a simple score (Impact × Confidence ÷ Effort). Create a 4–8 week editorial calendar with owners, publish dates, review checkpoints, and CTAs.
  • Outputs: Prioritized backlog + editorial calendar.
  • Checks: Calendar is feasible (no hidden approvals), and every item has an owner and a distribution plan.

6) Create content briefs (make writing easy)

  • Inputs: Top topics, voice rules, CTA, distribution channels.
  • Actions: Write 3 briefs using references/TEMPLATES.md: target query/intent, angle (“information gain”), outline, proof assets needed (SME quotes/data), CTA, internal/external links, distribution checklist.
  • Outputs: 3 content briefs (ready to draft).
  • Checks: A writer can draft without additional meetings; each brief includes what must be uniquely true (not generic).

7) Draft one flagship asset (blog post or announcement)

  • Inputs: One brief (or announcement context), references/TEMPLATES.md, references/WORKFLOW.md.
  • Actions: Draft an outline or first draft. If it’s an announcement, use a blog-post format with all “news” elements included. Use AI only per the SOP; add human insight, examples, and original framing.
  • Outputs: 1 draft outline/first draft + a “fact/claim check” list.
  • Checks: Draft has a clear POV, concrete examples, and avoids “100% AI-generated” sameness.

8) Measurement + quality gate + finalize

  • Inputs: Draft pack; references/CHECKLISTS.md and references/RUBRIC.md.
  • Actions: Define measurement cadence (weekly review) and iteration loop (what you’ll change based on signals). Run checklist and score rubric. Always include Risks / Open questions / Next steps.
  • Outputs: Final Content Marketing Plan Pack.
  • Checks: Next 2 weeks of execution are unblocked; assumptions and demand validation gaps are explicit.

Quality gate (required)

Examples

Example 1 (B2B SaaS, SEO + pipeline): “Use content-marketing. Product: AI note-taking for sales calls. ICP: SDR managers at 50–500 person SaaS. Goal: increase qualified demo requests; timebox: 8 weeks; constraints: 1 marketer + 2 SMEs, compliance review required. Output: a Content Marketing Plan Pack with demand-validated SEO topics, an editorial calendar, 3 briefs, and one drafted flagship post.”

Example 2 (Founder-led thought leadership): “Our founder wants to become the face of the brand on LinkedIn. Pick a primary format, define voice rules, build a 6-week content calendar, and create 3 briefs that address our audience’s career anxieties.”

Boundary example (upstream missing ICP): “We don’t know who we’re for yet. Create a content strategy anyway.” Response: pause and recommend defining ICP/positioning first; offer a minimal discovery sprint and then return to content planning.

平台分布

Codex

38.08%
按下载量换算94

Claude

31.45%
按下载量换算78

Cursor

18.97%
按下载量换算47

Gemini CLI

8.77%
按下载量换算22

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills