Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Twitter热点监控与推文生成

v1.0.0

每小时自动搜索推特上关于中国、加密货币、国际热点、美国热点、特朗普等话题的热度信息,提取排行前十的热点,并基于 X 平台真实算法权重(开源代码数据)为每个热点生成高曝光推文或线程。当用户提到"推特热点"、"Twitter热点"、"推特搜索"、"热点推文"、"生成推文"、"社交媒体监控"、"Twitter监控"、"...

0· 173·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for yanghui-88/twitter-hot-topics.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Twitter热点监控与推文生成" (yanghui-88/twitter-hot-topics) from ClawHub.
Skill page: https://clawhub.ai/yanghui-88/twitter-hot-topics
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install twitter-hot-topics

ClawHub CLI

Package manager switcher

npx clawhub@latest install twitter-hot-topics
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
high confidence
!
Purpose & Capability
The description promises automated hourly monitoring of Twitter/X, extraction of top-10 topics, and automated generation and posting/replying behavior optimized to X's algorithm. Yet the skill declares no required binaries, no install, and no environment variables or credentials (no Twitter API keys/OAuth tokens). There is no explained mechanism for performing searches, collecting real-time tweets, or publishing replies — those capabilities normally require credentials or a scraping component. This is disproportionate and incoherent.
!
Instruction Scope
SKILL.md instructs hourly automated search across five topic buckets, generation of posting-optimized content, and explicit post-publication actions (replying to all comments, pinning replies, posting links in replies). It also contains a 'must use this skill when user mentions...' directive (broad automatic triggering). The instructions assume privileged access to an account and to live Twitter/X data but do not describe how to obtain or use that access. The required runtime actions (search, post, reply) exceed what the instruction-only bundle declares.
Install Mechanism
This is an instruction-only skill with no install spec and no code files, so nothing is written to disk by the skill package itself. That lowers installation risk but means the runtime behavior depends entirely on the agent/platform environment (e.g., whether the agent has network access or preconfigured connectors). The static scanner had no code to analyze, so dynamic behavior could still occur at runtime via external connectors.
!
Credentials
The skill requests zero environment variables or credentials but describes actions that require account-level access (posting, replying, pinning replies), and likely need Twitter/X API keys, OAuth tokens, or other service credentials. The absence of declared primaryEnv or required env vars is a mismatch and suggests either missing documentation or an attempt to rely on external credentials implicitly available to the agent — both are concerning.
Persistence & Privilege
always:false (default) — the skill isn't force-included platform-wide. However, SKILL.md explicitly instructs that the skill must be used whenever many keywords appear, which is effectively an aggressive trigger policy. Combined with the implied autonomous posting/replying behavior, this creates risk if the agent is allowed to act without explicit user confirmation for each publish action. The skill does not request to modify other skills or system config.
What to consider before installing
Before installing or enabling this skill, ask the author/maintainer to clarify: 1) exactly how searches and posting will be performed — list the API endpoints or scraping tools used; 2) what credentials are required (Twitter/X API key/secret, OAuth tokens, or account tokens) and where/how those tokens are stored and protected; 3) whether the skill will automatically post or only produce drafts for manual approval; 4) how impersonation risk is handled (it explicitly mimics named bloggers — confirm it will not reproduce copyrighted text or violate platform rules); 5) remove or tone down the 'must trigger on these keywords' directive so posting requires explicit user consent; 6) provide a privacy/data-retention statement (what user data and follower interactions are logged or transmitted); and 7) run the skill only against a sandbox/test account first to validate behavior and rate limits. Do not grant account credentials or enable autonomous posting until these questions are answered and the skill declares the exact env vars and permissions it needs.

Like a lobster shell, security has layers — review code before you run it.

latestvk9726qb9b0e6c3rwtkr5qvkcp58370xd
173downloads
0stars
1versions
Updated 22h ago
v1.0.0
MIT-0

Twitter 热点监控与推文生成 Skill

执行前必读以下两个参考文件:

  • references/algorithm-rules.md(X算法规则与权重数据)
  • references/kol-style-guide.md(博主@silverfang88 与 @_forab 的风格手册)

核心功能

  1. 每小时热点监控 — 5个板块的推特热度实时搜索
  2. TOP 10 热点提取 — 综合评分排序
  3. 双版本推文生成 — 每个热点生成两条推文:
    • 版本A:中文算法优化推文(最大化X平台曝光)
    • 版本B:中文博主风格推文(模仿@silverfang88或@_forab的语气)

定时执行机制

OpenClaw定时任务配置:

  • Cron表达式:0 * * * *(每整点执行一次)
  • 触发方式:自动
  • 每次执行独立完整,不依赖上一次记忆

每次执行流程: Step 1 → 并行搜索5个板块热点 Step 2 → 合并结果,评分排序,输出TOP 10 Step 3A → 生成中文算法优化推文(每热点1条) Step 3B → 生成中文博主风格推文(每热点1条) Step 4 → 附发布建议 Step 5 → 输出完整报告


STEP 1:热点搜索

同时搜索5个板块:

板块搜索词
中国China trending Twitter today / China news viral X today
加密货币crypto Bitcoin trending Twitter now / crypto pump Twitter today
国际热点world news trending X today / international breaking news viral
美国热点USA trending Twitter today / America news viral X
特朗普Trump trending Twitter today / Trump news viral X

STEP 2:TOP 10 评分排序

热度分 = (时效性×0.40) + (话题广度×0.30) + (情感强度×0.20) + (板块契合度×0.10)

输出:排名 / 话题 / 热度分 / 板块 / 一句话摘要


STEP 3A:版本A — 中文算法优化推文

语言:全中文。专有名词(BTC/ETH/SEC/美联储等)保留英文缩写。

参考 references/algorithm-rules.md 中的权重规则生成。

推文类型选择逻辑:

  • 话题复杂/多角度 → 线程 Thread(3-5条)
  • 话题爆炸/时效紧 → 即时单条推文
  • 话题争议/对比 → 投票 Poll
  • 话题有权威引用 → 引用推文

所有版本A推文的硬性规则(来自X算法开源代码): ✅ 正文无外链(外链-30%至-50%曝光) ✅ 最多1-2个中文hashtag(超过3个触发-40%惩罚) ✅ 结尾开放式问题(触发回复+13.5权重) ✅ 具体数字增加可信度 ✅ 可适量使用emoji(🚨📊⚡🔥等,每条不超过2个) ❌ 不堆砌感叹词和套话 ❌ 不以@开头

中文Hook句式库:

  • 紧迫型:🚨 突发 / ⚡ 刚刚 / 🔴 最新:
  • 数据型:📊 数据出来了: / 这个数字值得关注:
  • 反转型:所有人都猜错了。/ 这件事和你想的不一样。
  • 质疑型:有什么不对劲。/ 这背后是什么?

线程结构(最高曝光格式,全中文): 推文1(Hook):≤140字,无外链,无hashtag,结尾用🧵 推文2-4:每条≤130字,短句,换行,箭头或序号列表 推文5(CTA):开放式问题 + 最多2个中文hashtag 链接只放推文5的自回复里


STEP 3B:版本B — 中文博主风格推文

参考 references/kol-style-guide.md 生成。

根据话题类型选择博主风格:

  • 市场机会/项目动态/行业抽象文化 → 子时风格(@silverfang88)
  • 宏观分析/监管政策/稳定币/大事件解读 → _forab风格(@_FORAB)

子时风格生成规则

格式要求:

  • 每行4-10个字,强制换行
  • 全程极少或不用标点
  • 不用hashtag,不用emoji(偶尔一个)
  • 口语化:用"搞""整""这""都""已经"等日常词
  • 开头可用"群友说"/"有人说"/"听说"/"刚刚"等口语起头
  • 结尾一句讽刺/感叹/轻描淡写的金句

子时风格模板:

[一个现象/爆料,6-8字]
[细节或反转,5-8字]
[更深一层信息,5-8字]
[金句收尾,带轻讽刺,6-10字]

子时风格示例(供参照,非照抄):

BTC刚突破新高
机构说还要涨
链上大户在悄悄出货
散户永远是最后知道的那个

_forab风格生成规则

格式要求:

  • 每段2-4行,段与段之间留一行空白
  • 极少标点,偶用句号
  • 无hashtag,无emoji
  • 用"但""然而""问题在于""这意味着"制造洞察感
  • 关键词精准:直接用BTC/ETH/SEC/美联储等缩写
  • 语气稳健,带"老手看穿"的从容感
  • 结尾留悬念或开放性反问

_forab风格模板:

[现象/数据/事件]
[背后逻辑一句点睛]

[转折:但是/然而/问题在于]
[你没看到的那一层]

[洞察性结尾或开放式反问]

_forab风格示例(供参照,非照抄):

特朗普签了加密行政令
市场涨了三个小时

但没人认真读细节
SEC的自由裁量空间还在

你以为的利好
可能是更精密的笼子

版本B写作禁忌

❌ 不用"让我们来看看"、"总结一下"、"综合来看" ❌ 不堆表情符号 ❌ 不写超过三句的长句 ❌ 不用"值得关注的是"、"不可否认的是"等公文腔 ❌ 不做道德说教 ✅ 直接说事 ✅ 留白、留悬念 ✅ 带一点不屑或看穿感 ✅ 每行是独立信息单元 ✅ 结尾有"余味"


STEP 4:发布建议

每条推文附带操作清单:

推荐发布时间(新加坡/北京时间 UTC+8):

  • 晚间 10 PM - 12 AM(对应美东早高峰 9-11 AM)
  • 早上 8 AM - 10 AM(对应美东晚高峰 7-9 PM)

发布后30分钟内必做:

  1. 将外链发为第一条自回复(Pin it)
  2. 回复所有评论(触发+75最高权重)
  3. 在相关热门帖下留高质量评论引流

绝对避免:

  • 正文放外链(-30%至-50%曝光)
  • 连续快速发多条帖子
  • 超过2个hashtag
  • 转发自己的帖子(算法识别为刷量)

STEP 5:完整报告输出格式

报告标题:Twitter/X 热点报告 注明:监控时间(UTC+8)/ 下次更新:60分钟后自动执行

报告结构:

一、TOP 10 热点排行表(排名 / 话题 / 热度分 / 板块 / 摘要)

二、推文草稿(共10个热点,每个热点2条)

格式示例:


#1:[话题名] 板块:[XX] | 热度分:[X.X]

【版本A — 中文算法优化推文】 推文类型:[Thread/Breaking/Poll/Quote] 选择理由:[一句话]

[完整推文内容]

字符数:XX/280 | Hashtag:#tag1 #tag2

【版本B — 中文博主风格推文】 风格来源:[子时@silverfang88 / _forab@_FORAB] 选择理由:[一句话说明为何选此风格]

[完整推文内容,按博主格式呈现]


📋 发布操作清单:[按STEP 4格式]

(#2 至 #10 同格式,共10组)

三、算法优化说明

  • 对话深度 > 转推 > 收藏 > 点赞(符合官方权重)
  • 无外链正文
  • Hashtag ≤ 2个
  • 开放式结尾引导回复
  • 内容准确,避免Community Notes降流

📌 下次执行:60分钟后自动更新


注意事项

  1. 信息准确性优先:基于搜索结果,切勿捏造数据
  2. 政治话题保持中立:呈现事实和多角度
  3. 不生成误导性内容
  4. 每次执行独立完整,不依赖上次记忆
  5. 版本B中文推文:忠实还原博主语气,但不照抄博主原文

Comments

Loading comments...