Practical News Monitor

支持地缘政治、石油、黄金等关键词的多数据源新闻监控，自动保存数据并生成摘要报告，易于扩展和定制。

Security Scan

The skill is a straightforward news-monitoring script that fetches public news pages/APIs, parses items, and saves JSON/markdown reports locally; its requirements and behavior align with that purpose.

Purpose & Capability

The name/description (news monitor for geopolitics, oil, gold) match the included Python script and SKILL.md. The code implements JSON API and HTML parsing sources, keyword matching, and local persistence — all expected for this functionality.

Instruction Scope

Instructions are limited to installing requests, running the Python script, editing keyword/source configuration, and reading/writing data in a local DATA_DIR. This is within scope. Note: the skill will perform network requests to whatever source URLs are configured and will write files to the user's filesystem (default: ~/shared_memory/practical_news). Users should review configured source URLs before running.

Install Mechanism

No install spec is provided (instruction-only + a code file). SKILL.md advises pip3 install requests — expected for the script. There is no remote download of arbitrary binaries or extract steps. The lack of pinned package versions is a minor maintenance/security note but not an incoherence.

Credentials

The skill requests no environment variables or credentials and does not reference system config paths beyond a user-writable data directory. This is proportional to a news-monitoring tool.

Persistence & Privilege

The skill is not always-enabled, does not request elevated privileges, and only persists its own data files in a configurable local directory. It does not modify other skills or agent-wide configuration in the provided files.

Assessment

This appears to be a benign, self-contained news-monitoring script, but before installing/running: 1) Review and update the source list in _init_sources() so it only queries trusted public endpoints; 2) Change DATA_DIR to a directory you control (avoid running as a privileged user); 3) Inspect the remainder of the Python file (the provided file was truncated in the manifest) to confirm there are no hidden network endpoints or telemetry; 4) Consider pinning the requests dependency (e.g., requests==2.x) and running in a virtualenv; 5) If you plan to run it regularly, run it in a restricted environment or sandbox and monitor its network activity. If you want, I can scan the rest of the file (the truncated portion) or highlight exact lines that write files or perform network requests.

---

name: practical-news-monitor description: 实用新闻监控框架 - 专注于可工作的数据源和易扩展性，支持地缘政治、石油、黄金等关键词监控 metadata: {"clawdbot":{"emoji":"📰","requires":{"bins":["python3","requests"]}}}

实用新闻监控框架

专注于可工作的数据源和易扩展性，支持地缘政治、石油、黄金等关键词监控。

核心特性

• 📡 多数据源支持: JSON API 和 HTML 解析两种数据源类型
• 🎯 关键词监控: 可自定义监控关键词和分类
• 💾 数据持久化: 自动保存监控数据到 JSON 文件
• 📊 摘要报告: 自动生成新闻监控摘要报告
• 🔧 易扩展: 基类设计，轻松添加新数据源

快速开始

安装依赖

pip3 install requests

运行监控

cd /path/to/skill
python3 practical_news_monitor.py monitor

测试数据源

python3 practical_news_monitor.py test

配置说明

监控关键词

在脚本中编辑 MONITOR_KEYWORDS 字典：

MONITOR_KEYWORDS = {
    'geopolitical': ['中东', '伊朗', '以色列', '巴勒斯坦'],
    'oil': ['石油', '原油', '油价', 'OPEC'],
    'gold': ['黄金', '贵金属', '避险', '金价'],
    'sanctions': ['制裁', '禁令', '限制', '封锁'],
    'shipping': ['航运', '海运', '物流', '苏伊士'],
    'conflict': ['冲突', '战争', '打击', '攻击']
}

添加数据源

在 _init_sources() 方法中添加新的数据源配置：

source_configs = [
    {
        'type': 'json_api',
        'name': '我的JSON API',
        'config': {
            'enabled': True,
            'url': 'https://api.example.com/news',
            'data_path': 'data.items',
            'title_field': 'title',
            'url_field': 'link',
            'time_field': 'published_at',
            'timeout': 15,
            'limit': 20
        }
    },
    {
        'type': 'html_parse',
        'name': '我的HTML网站',
        'config': {
            'enabled': True,
            'url': 'https://example.com/news',
            'parse_rules': {
                'container_selector': r'<div class="news-item">(.*?)</div>',
                'title_selector': r'<a[^>]*>([^<]+)</a>',
                'link_selector': r'href="([^"]+)"'
            },
            'timeout': 15,
            'limit': 20
        }
    }
]

数据目录

默认数据保存位置：

/Users/zenozhou/shared_memory/practical_news/

可以在脚本中修改 DATA_DIR 变量来更改位置。

输出示例

监控运行后会生成：

1. JSON 数据文件: 包含所有新闻和相关性信息
2. 摘要报告: Markdown 格式的新闻摘要

============================================================
📊 实用新闻监控
============================================================
监控时间: 2026-03-23 08:30:00

📡 正在获取 新浪财经（示例）...
   ✓ 获取到 20 条新闻
📡 正在获取 东方财富（示例）...
   ✓ 获取到 15 条新闻

✅ 共获取 35 条新闻
📈 相关新闻: 8 条

📊 来源统计:
  新浪财经（示例）: 5/20 相关
  东方财富（示例）: 3/15 相关

💾 数据已保存: /Users/zenozhou/shared_memory/practical_news/monitor_2026-03-23_083000.json

扩展开发

创建新数据源类型

继承 NewsDataSource 基类并实现 fetch() 方法：

class MyCustomSource(NewsDataSource):
    """我的自定义数据源"""

    def fetch(self) -> List[Dict[str, Any]]:
        """获取新闻数据"""
        # 实现你的数据获取逻辑
        pass

自定义相关性检查

重写 check_relevance() 方法：

def check_relevance(self, title: str, keywords: Dict[str, List[str]]) -> Dict[str, Any]:
    """自定义相关性检查逻辑"""
    # 实现你的相关性检查
    pass

常见问题

Q: 如何添加新的关键词分类？

A: 在 MONITOR_KEYWORDS 字典中添加新的键值对即可。

Q: 数据源返回 HTTP 错误怎么办？

A: 检查 URL 是否正确，是否需要 headers 或 cookies，尝试在浏览器中访问确认。

Q: 如何调整监控频率？

A: 使用 cron 或其他调度工具定期运行脚本。

许可证

MIT License

贡献

欢迎提交 Issue 和 Pull Request！

---

由 Zbot 自动生成 💥