Trending News Aggregator

智能热点新闻聚合器 - 自动抓取多平台热点新闻， AI分析趋势，支持定时推送和热度评分。 核心功能： - 每天自动聚合多平台热点（微博、知乎、百度等） - 智能分类（科技、财经、社会、国际等） - 热度评分算法 - 增量检测（标记新增热点） - AI趋势分析

Security Scan

The skill's description and runtime instructions mostly match an aggregator, but there are notable inconsistencies around installation and credential handling (push channels) that need clarification before trusting it.

Purpose & Capability

The skill's name/description align with the instructions: it uses web_search to aggregate and classify trending items. Declared dependency on a 'web_search' tool is appropriate. However, package.json and PUBLISH.md reference an npm package and GitHub repo (placeholders) even though this bundle is instruction-only and contains no runnable source — that's an inconsistency in how the skill would be installed or distributed.

Instruction Scope

SKILL.md confines runtime behavior to web searches, classification, scoring, and optional pushes; it does not instruct reading unrelated system files or secrets. It does advise cloning a GitHub repo and copying into ~/.openclaw/skills for manual install, which is standard but points to an external repo that is a placeholder and not verifiable here.

Install Mechanism

Registry metadata says 'no install spec' (instruction-only), which is low-risk. But package.json includes an openclaw.install.npmSpec and PUBLISH.md describes publishing to npm/ClawHub — conflicting signals about how to obtain code. There is no actual code to be installed in this package, so the install/publish instructions are incomplete or misleading.

Credentials

The skill declares no required environment variables or credentials, yet supports multiple push channels (Weixin, DingTalk, Slack, Telegram, SMTP) that necessarily require tokens/credentials to operate. SKILL.md does not declare or document expected env var names or secure credential storage/config paths. This omission creates a risk that users will be asked later to provide secrets in an ad hoc way or configure them unsafely.

Persistence & Privilege

always is false and there is no request for system-wide privileges. The skill suggests storing configuration locally (config.yaml) and scheduling tasks (cron/OpenClaw cron), which is expected behavior for an automation skill and not excessive.

What to consider before installing

Key things to verify before installing: - Credentials for push channels: ask the author which env vars or config fields are required and how secrets are stored; do not paste tokens into chat or public fields. The skill currently lists push targets but declares no credential names. - Source verification: the SKILL references cloning a GitHub repo and npm publishing, but the repo URL and npm spec are placeholders (yourusername). Prefer skills with a real, auditable repository and published package rather than anonymous instruction-only bundles. - Installation ambiguity: package.json suggests an npm install path though the package here has no executable code — clarify whether code exists elsewhere and where it will be downloaded from. - Test safely: run the skill in a sandboxed environment first (no real push credentials) and review any code from the claimed repo before enabling scheduled pushes. - If you need automated push notifications, require explicit documentation of credential names, storage location, and least-privilege instructions; consider using per-channel bot accounts or webhooks you control and revokeable tokens. If the maintainer cannot provide a real repository and clear credential handling, treat this skill as untrusted.

智能热点新闻聚合器，自动抓取多平台热点，AI分析趋势。

核心功能

综合热点聚合 (Trending News)

• 频率: 可配置，建议每天9:00
• 内容: 多平台热点聚合 + 智能分类 + AI趋势分析
• 输出: 分类热点 + 热度评分 + 增量标记 + 趋势总结

智能分类系统

• 科技: AI、芯片、华为、字节、腾讯等
• 财经: 股市、经济、金融、投资等
• 社会: 民生、教育、医疗、就业等
• 国际: 美国、中国、欧洲、印度等
• 娱乐: 电影、明星、综艺、音乐等

热度评分算法

基于以下维度加权计算：

• 平台排名权重
• 多平台出现频次
• 新闻时效性
• 用户互动数据

使用方法

手动触发

获取今日热点:

获取今日热点新闻

或详细指令：

【综合热点聚合】执行以下步骤：
1) 使用web_search搜索各平台热点新闻
2) 关键词智能分类：科技/财经/社会/国际/娱乐
3) 热度计算：按排名+频次+时效加权
4) 增量检测：标记新增热点
5) AI趋势分析：一句话总结今日热点趋势
6) 输出格式：总新闻：X条 | 新增：X条 | 时间：XX:XX
   🔥高热度分类：[1/X]分类名（X条）[热度分]
   每条新闻格式：1.[平台] 标题 [排名][🔥热度][🆕新增] 链接：https://具体新闻页面链接

定时任务配置

方式1：使用OpenClaw Cron

# 每天9:00执行
schedule: "0 9 * * *"

方式2：系统定时任务

# Linux/Mac crontab
0 9 * * * openclaw run-task trending-news

# Windows Task Scheduler
# 每天 9:00 执行

推送渠道配置

支持多种推送渠道（需单独配置）：

• 微信 (weixin): 个人号/群聊推送
• 钉钉 (dingtalk): 群机器人推送
• Slack: 频道消息
• Telegram: 频道/群组
• 邮件: SMTP发送

配置示例：

push:
  enabled: true
  channels:
    - weixin
    # - dingtalk
    # - slack

输出格式示例

===今日热点=== 2026-03-25 09:00
总新闻：32条 | 新增：8条

🔥高热度分类：
[1/5] 科技（12条）[热度分：95]
  1.[微博] 华为发布新一代AI芯片 [🔥98][🆕新增]
    链接：https://weibo.com/xxx
  2.[知乎] GPT-5技术细节曝光 [🔥92]
    链接：https://zhihu.com/xxx
  3.[百度] 国产量子计算机突破 [🔥88][🆕新增]
    链接：https://baidu.com/xxx

[2/5] 财经（8条）[热度分：88]
  1.[新浪财经] A股大涨 [🔥85]
    链接：https://sina.com.cn/xxx
  ...

[3/5] 国际（6条）[热度分：76]
  ...

📈 AI趋势分析：
今日科技板块热度最高，AI和芯片成为焦点，
反映市场对技术创新的高度关注。

配置说明

config.yaml

# 新闻源配置
sources:
  weibo: true      # 微博热搜
  zhihu: true      # 知乎热榜
  baidu: true      # 百度热搜
  douyin: true     # 抖音热点
  toutiao: false   # 今日头条（按需启用）

# 智能分类关键词
categories:
  tech:
    name: "科技"
    keywords: ["AI", "人工智能", "芯片", "量子", "华为", "字节", "腾讯", "阿里", "百度", "小米"]
    
  finance:
    name: "财经"
    keywords: ["股市", "A股", "港股", "美股", "经济", "金融", "银行", "投资", "基金"]
    
  social:
    name: "社会"
    keywords: ["社会", "民生", "教育", "医疗", "就业", "房价", "养老", "环保"]
    
  international:
    name: "国际"
    keywords: ["美国", "中国", "日本", "韩国", "欧洲", "印度", "俄罗斯", "联合国"]
    
  entertainment:
    name: "娱乐"
    keywords: ["电影", "电视剧", "明星", "综艺", "音乐", "演唱会", "票房"]

# 热度评分配置
scoring:
  base_score: 50
  rank_weight: 10
  frequency_weight: 5
  freshness_weight: 15
  max_score: 100

# 推送设置
push:
  enabled: true
  channels:
    - weixin
  schedule: "09:00"

# 输出格式
output:
  show_category_count: true
  show_heat_score: true
  show_new_mark: true
  show_trend_analysis: true
  max_news_per_category: 5

依赖

必需:

• web_search tool (Brave Search / Tavily)
• OpenClaw >= 2026.3.22

可选:

• 推送渠道配置（微信/钉钉/Slack等）

安装

通过 ClawHub 安装（推荐）

clawdhub install trending-news-aggregator

手动安装

# 1. 克隆仓库
git clone https://github.com/yourusername/trending-news-aggregator.git

# 2. 复制到OpenClaw skills目录
cp -r trending-news-aggregator ~/.openclaw/skills/

# 3. 重启OpenClaw或刷新skills

使用示例

示例1：手动获取热点

用户: 今天有什么热点新闻？

Agent:

我来为您获取今日热点新闻...

===今日热点=== 2026-03-25 09:00
总新闻：28条 | 新增：6条

🔥高热度分类：
[1/4] 科技（10条）[热度分：92]
  1.[微博] 华为发布新芯片 [🔥95][🆕新增]
  2.[知乎] AI技术突破 [🔥88]
  ...

📈 趋势分析：科技板块持续火热

示例2：设置定时推送

用户: 每天早上9点给我推送新闻

Agent:

已为您设置每日9:00热点推送。

配置详情：
- 任务：综合热点聚合
- 时间：每天 09:00
- 渠道：微信
- 内容：科技/财经/社会/国际热点

示例3：查看配置

用户: 新闻聚合器配置了哪些分类？

Agent:

当前配置的分类：

1. 科技 - AI、芯片、华为、字节等
2. 财经 - 股市、经济、金融等
3. 社会 - 民生、教育、医疗等
4. 国际 - 美国、中国、欧洲等
5. 娱乐 - 电影、明星、综艺等

共5个分类，覆盖32个关键词。

隐私说明

本Skill：

• ✅ 不收集用户个人信息
• ✅ 不存储浏览历史
• ✅ 仅使用公开网络搜索
• ✅ 所有配置存储在用户本地

注意事项

1. API限制: web_search有调用频率限制，请合理使用
2. 新闻时效: 新闻链接有效期有限，建议及时查看
3. 分类准确性: AI分类可能存在偏差，仅供参考
4. 推送测试: 首次使用建议先手动测试推送渠道

故障排除

问题1：无法获取新闻

• 检查web_search工具是否启用
• 检查网络连接
• 查看OpenClaw日志

问题2：推送失败

• 检查推送渠道配置
• 确认微信/钉钉等已正确配置
• 查看推送渠道日志

问题3：分类不准确

• 在config.yaml中调整关键词
• 添加或删除分类
• 反馈给开发者改进算法

更新日志

v1.0.0 (2026-03-25)

• 初始版本发布
• 多平台热点聚合功能
• 智能分类系统
• 热度评分算法
• 增量检测功能
• AI趋势分析
• 定时推送支持

贡献

欢迎提交Issue和PR！

GitHub: https://github.com/yourusername/trending-news-aggregator

License

MIT License

Copyright (c) 2026 OpenClaw Community