← 返回 Skills 市场
peterdog666

热点雷达

作者 Peter Dog · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
93
总下载
1
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install hotspot-radar
功能描述
全网热榜追踪器,聚合微博/知乎/抖音/B站/小红书热搜榜单,支持趋势分析、话题监控和定时推送 (user)
使用说明 (SKILL.md)

\r \r

热点雷达 - 全网热榜追踪器\r

\r 你是一个全网热点信息追踪专家,能够实时聚合微博、知乎、抖音、B站、小红书五大平台的热搜榜单,为用户提供一站式的热点发现与分析服务。\r \r

核心能力\r

\r

1. 实时热榜获取\r

  • 微博热搜榜\r
  • 知乎热榜\r
  • 抖音热搜榜\r
  • B站热榜\r
  • 小红书热榜\r \r

2. 趋势追踪分析\r

  • 对比昨日热榜,发现新上榜/落榜话题\r
  • 计算话题热度变化趋势\r
  • 识别"正在起飞"的热点\r \r

3. 话题监控\r

  • 用户设定关键词监控(如"AI"、"考研"、"新能源")\r
  • 当监控话题出现在热榜时及时提醒\r
  • 支持多关键词同时监控\r \r

4. 定时报告推送\r

  • 每日早报(9:00)和晚报(21:00)\r
  • 自动生成Markdown格式热点报告\r
  • 支持推送到飞书群\r \r

5. HTML可视化报告\r

  • 生成交互式HTML报告(深色主题)\r
  • 内嵌Chart.js图表:各平台热度对比柱状图、新增/上升话题横向条形图\r
  • 平台热榜卡片式展示,支持TOP10详情\r
  • 跨平台热点聚合展示\r
  • 话题监控告警高亮卡片\r
  • 命令:node scripts/index.js html\r \r

标准工作流程\r

\r

获取全网热榜\r

  1. 并行调用各平台数据采集接口\r
  2. 统一数据格式,按热度排序\r
  3. 生成平台对比表格\r
  4. 输出Markdown格式热榜报告\r \r

趋势分析模式\r

  1. 读取历史热榜数据(存储在 data/history/ 目录)\r
  2. 与当前热榜对比\r
  3. 标注:新上榜(↑NEW)、排名上升(↑)、排名下降(↓)、落榜(↓OUT)\r
  4. 生成趋势变化摘要\r \r

话题监控模式\r

  1. 读取监控配置(config/monitor.json)\r
  2. 在当前热榜中搜索匹配话题\r
  3. 如有匹配,生成提醒报告\r
  4. 更新监控日志\r \r

生成报告\r

报告格式:\r

# 🌐 全网热点日报 | {日期}\r
\r
## 今日概览\r
- 微博热搜:{条数}条 | 知乎热榜:{条数}条 | 抖音热搜:{条数}条\r
- B站热榜:{条数}条 | 小红书热榜:{条数}条\r
- 发现 {n} 个跨平台热点话题\r
\r
## 跨平台热点 TOP5\r
1. [话题名称] - 出现在微博、知乎、抖音\r
2. ...\r
\r
## 各平台热榜\r
### 微博热搜\r
| 排名 | 话题 | 热度 |\r
|------|------|------|\r
| 1 | xxx | 1234万 |\r
\r
... (其他平台同理)\r
\r
## 趋势变化\r
### 🆕 新上榜\r
- [话题] - 首度登榜\r
\r
### 📈 热度上升\r
- [话题] - 从第X名升至第X名\r
\r
### 📉 落榜\r
- [话题] - 昨日第X名,今日跌出榜单\r
\r
## 话题监控提醒\r
- [关键词]: 出现在微博第X位、知乎第X位\r
\r
---\r
生成时间: {timestamp}\r
```\r
\r
## 数据存储\r
\r
- `data/history/{platform}/{YYYY-MM-DD}.json` - 各平台历史热榜\r
- `data/trends/{YYYY-MM-DD}.json` - 每日趋势分析结果\r
- `config/monitor.json` - 话题监控配置\r
- `config/push.json` - 推送渠道配置\r
\r
## 错误处理\r
\r
- 某平台获取失败时,跳过该平台并标注"数据获取失败"\r
- 不影响其他平台正常展示\r
- 自动重试机制:首次失败后等待3秒重试,最多3次\r
\r
## 注意事项\r
\r
- 尊重平台数据版权,仅展示话题名称,不深度爬取详细内容\r
- 定时任务需要用户确认推送渠道配置\r
- 首次使用需运行初始化脚本配置数据目录\r
安全使用建议
This skill appears to do what it claims — aggregate public hot-topic lists, analyze trends, and generate reports — but check a few things before enabling it: - Third-party endpoints: the collector queries third-party aggregator APIs (apiserver.alcex.cn and 60s.viki.moe) in addition to a Weibo endpoint. Those services supply the data; if you don't trust them, consider replacing or auditing those endpoints. The skill makes outbound GET requests only (no local data is sent to those servers). - Push/webhook configuration is optional but powerful: config/push.json contains placeholders for Feishu/webhook and email. Do not populate those fields with sensitive credentials unless you trust the skill and its maintainer. - Reports and HTML: reports and the HTML report load remote resources (Chart.js via CDN). Opening a generated HTML file will cause your browser to fetch resources from the CDN. Also, while the code serializes data as JSON into the page (which is normally safe), exercise caution if you plan to publish or share generated HTML without sanitizing platform-provided topic text. - Code provenance: there is no homepage or maintainer identity. If you require higher assurance, run the scripts in an isolated environment (container or VM), or review/replace the external API endpoints with sources you trust. - Minor inconsistency: README mentions optional cookies via env vars but the scripts don't actually use environment variables; this is not harmful but worth noting. If you accept those caveats, the skill is coherent and operates within its described scope.
功能分析
Type: OpenClaw Skill Name: hotspot-radar Version: 1.0.0 The hotspot-radar skill is a legitimate utility designed to aggregate and analyze trending topics from platforms like Weibo, Zhihu, and Bilibili. The codebase (scripts/collector.js, scripts/htmlReporter.js, etc.) uses standard Node.js modules to fetch data from public third-party APIs and generate local Markdown and HTML reports. The requested permissions, including ExecuteCommand and WebFetch, are strictly aligned with its stated purpose of running internal reporting scripts and gathering data. No evidence of data exfiltration, malicious command execution, or prompt injection was found.
能力评估
Purpose & Capability
Name/description match the code: the scripts fetch hot lists from multiple platforms, compute trends, support keyword monitoring and generate Markdown/HTML reports. The files present (collector, reporter, htmlReporter, monitor, index) are appropriate for the described functionality. One minor inconsistency: README documents optional WEIBO_COOKIE/XHS_COOKIE env vars, but the code does not read process.env for these cookies.
Instruction Scope
SKILL.md and scripts limit activity to fetching platform APIs, reading/writing under data/ and config/, and generating reports. The agent is instructed to read data/history and config/monitor.json which matches code behavior. Note: generated HTML loads Chart.js from a CDN and embeds platform/topic data into the HTML/JS; this is expected for visualization but means opening the report will make the browser fetch remote resources and render data coming from external APIs.
Install Mechanism
There is no install spec — the skill is instruction/code only and runs with Node (no dependencies in package.json). That minimizes install risk. The code will be written to disk as part of the skill bundle but there is no external archive or untrusted installer.
Credentials
The skill requests no required environment variables or credentials in the registry metadata. It does include a push configuration (config/push.json) with an optional Feishu webhook field that is empty by default — enabling push would require the user to supply credentials. The code itself does not access other secrets. The only noteworthy point: the README mentions optional WEIBO_COOKIE and XHS_COOKIE env vars for improved API limits, but the provided scripts do not actually read those env vars.
Persistence & Privilege
always:false and user-invocable:true. The skill writes only to its own data/ and config/ directories and does not modify other skill configurations or system-wide settings. No elevated or persistent platform privileges are requested.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install hotspot-radar
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /hotspot-radar 触发
  4. 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
v1.0.0
hotspot-radar 1.0.0 - 全网热榜追踪器首发: - 支持实时聚合微博、知乎、抖音、B站、小红书五大平台热搜榜单 - 提供趋势分析:对比历史数据,标注新上榜、上升、下降及落榜话题 - 支持话题关键词监控与及时提醒,多关键词同时检测 -支持Markdown和交互式HTML报告生成 - 报告内含热榜对比、跨平台热点TOP5、趋势变化和话题监控提醒 - 完善的数据存储方案与错误处理机制
元数据
Slug hotspot-radar
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

热点雷达 是什么?

全网热榜追踪器,聚合微博/知乎/抖音/B站/小红书热搜榜单,支持趋势分析、话题监控和定时推送 (user). 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 93 次。

如何安装 热点雷达?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install hotspot-radar」即可一键安装,无需额外配置。

热点雷达 是免费的吗?

是的,热点雷达 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

热点雷达 支持哪些平台?

热点雷达 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 热点雷达?

由 Peter Dog(@peterdog666)开发并维护,当前版本 v1.0.0。

💬 留言讨论