← 返回 Skills 市场
255
总下载
0
收藏
1
当前安装
2
版本数
在 OpenClaw 中安装
/install article-analyzer
功能描述
爆款文章拆解分析工具。当用户发来公众号/网页链接并要求拆解、分析、学习、拆文时触发。支持结构分析、金句提取、选题逻辑、情绪曲线、复刻建议。关键词:拆解、分析、拆文、学习这篇文章、爆文分析。
使用说明 (SKILL.md)
优化日期: 2026-04-02
爆文拆解分析
触发条件
- 用户发来公众号/网页链接 + "拆解""分析""拆文""学习这篇文章"等关键词
- 用户说"分析这篇爆文""帮我拆解这篇文章结构"
工作流
Step 1: 抓取全文
# 微信公众号文章:用 browser 工具(web_fetch 反爬)
browser open url=\x3C用户链接>
browser snapshot format=ai maxChars=15000
- 微信文章 必须用 browser,不用 web_fetch(反爬)
- 其他网站可用 web_fetch 或 browser 均可
- 目标:获取完整正文内容
Step 2: 拆解分析
对抓取到的全文执行以下分析维度:
2.1 结构分析
- 标题拆解(标题用了什么技巧?数字?疑问?反差?痛点?)
- 开头手法(故事?数据?提问?场景?共鸣?)
- 段落逻辑(总分总?递进?并列?对比?故事线?)
- 结尾方式(金句收尾?行动号召?开放问题?反转?)
2.2 金句提取
- 提取5-10条最有记忆点的句子
- 标注金句类型:比喻/对比/数据/金句体/反常识
2.3 选题逻辑
- 目标读者画像(是谁?痛点是什么?)
- 情绪曲线(开头什么情绪→中间怎么推→结尾什么情绪)
- 选题角度(是什么让它值得写?独家信息?新视角?争议点?实用价值?)
2.4 可复刻点
- 这个标题模式我可以怎么用?
- 开头手法可以套用在什么主题上?
- 哪些金句可以改编?
- 整体结构适合写什么类型的内容?
Step 3: 输出报告
默认直接在聊天中输出。如用户要求写文档:
# 创建飞书文档
feishu_doc create title=\x3C标题> content=\x3CMarkdown格式的分析报告>
报告模板
# 🔪 爆文拆解:{文章标题}
## 📊 基本信息
- 标题:{原标题}
- 来源:{公众号名}
- 字数:{估算}
- 阅读量:{如果能获取}
## 🏗️ 结构分析
- 标题技巧:{分析}
- 开头手法:{分析}
- 段落逻辑:{分析}
- 结尾方式:{分析}
## ✍️ 金句TOP5
1. "{金句}" — {类型标注}
2. ...
## 🎯 选题逻辑
- 目标读者:{画像}
- 情绪曲线:{描述}
- 选题价值:{为什么这篇能火}
## 💡 复刻建议
- 标题复刻:{套用模板}
- 开头复刻:{适用场景}
- 结构复刻:{适合什么内容类型}
- 金句改编:{改编示例}
快速模式
如果用户只说"快速拆解"或"简单看看",跳过详细分析,只输出:
- 标题技巧(一句话)
- 文章结构(一句话)
- 金句TOP3
- 能学什么(一句话)
安全使用建议
This skill appears coherent and low-risk: it fetches article text (using the agent's browser/web_fetch tools), analyzes it, and can optionally create a Feishu document. Before installing, note that: 1) any article URL or content you provide will be sent to the agent/tools for processing (avoid private/confidential links unless you trust the agent and its integrations); 2) the Feishu document step likely requires a Feishu integration or API token managed by your platform — verify what permissions that integration has; 3) the browser snapshot is limited (maxChars=15000) and may truncate very long articles, so provide full text if needed. If you want stricter controls, test the skill with public articles first and confirm platform-managed tool credentials and storage/retention policies.
功能分析
Type: OpenClaw Skill
Name: article-analyzer
Version: 1.0.1
The 'article-analyzer' skill is a legitimate tool designed to scrape, analyze, and summarize web articles (specifically WeChat) and optionally save reports to Feishu. The workflow in SKILL.md uses standard browser and document tools for their intended purposes, and the reference data in patterns.md is purely informational. No evidence of data exfiltration, unauthorized command execution, or malicious prompt injection was found.
能力评估
Purpose & Capability
Name/description (爆款文章拆解) match the instructions: fetch article text, perform structure/quote/angle analysis, and produce a report. The skill is instruction-only and uses agent tools (browser, web_fetch, feishu_doc) which are appropriate for the stated purpose.
Instruction Scope
SKILL.md limits runtime actions to fetching the provided URL (browser/web_fetch), analyzing the retrieved text across explicit dimensions, and optionally creating a Feishu doc. It does not instruct the agent to read unrelated files, pull arbitrary environment variables, or send data to unknown endpoints. Note: it will send user-supplied article content to the agent/tooling for analysis (expected for this purpose).
Install Mechanism
No install spec and no code files — lowest-risk instruction-only skill. Nothing is downloaded or written to disk by the skill itself.
Credentials
SKILL.md invokes tools like feishu_doc create but the skill declares no required environment variables. This is not necessarily malicious (platform-managed integrations often provide credentials), but the skill should document that creating a Feishu document requires a Feishu integration/credentials managed by the agent environment. Confirm Feishu/browser/web_fetch tools won't leak sensitive tokens or broad workspace access.
Persistence & Privilege
always:false and no persistent presence or changes to other skills/configs. The skill does not request system-wide privileges or attempt to modify other skill settings.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install article-analyzer - 安装完成后,直接呼叫该 Skill 的名称或使用
/article-analyzer触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.1
- Improved and clarified the skill description to better outline supported triggers and functionalities.
- Updated keywords in the description for enhanced discoverability and trigger accuracy.
- No changes to core workflow, analysis dimensions, or reporting template.
- Documentation now notes the optimization date (2026-04-02).
- No functional or structural changes to the analysis process or user workflow.
v1.0.0
- Initial release of the article-analyzer skill.
- Automatically analyzes and deconstructs viral articles from WeChat and web links.
- Provides structure breakdown, key quote extraction, topic logic, emotional curve, and actionable suggestions.
- Supports outputting analysis as a Feishu document if requested.
- Includes both detailed and quick analysis modes based on user prompt.
元数据
常见问题
Article Analyzer 是什么?
爆款文章拆解分析工具。当用户发来公众号/网页链接并要求拆解、分析、学习、拆文时触发。支持结构分析、金句提取、选题逻辑、情绪曲线、复刻建议。关键词:拆解、分析、拆文、学习这篇文章、爆文分析。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 255 次。
如何安装 Article Analyzer?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install article-analyzer」即可一键安装,无需额外配置。
Article Analyzer 是免费的吗?
是的,Article Analyzer 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Article Analyzer 支持哪些平台?
Article Analyzer 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Article Analyzer?
由 luaqnyin(@luaqnyin)开发并维护,当前版本 v1.0.1。
推荐 Skills