News Aggregator Pro

国内外社会、科技、军事新闻汇总。自动搜索、筛选、整理新闻要点。

Security Scan

The skill is a straightforward instruction-only news aggregator, but inconsistent metadata (owner/slug/version) and unknown origin reduce trust and warrant caution before installing.

Purpose & Capability

The name, description, and SKILL.md content are coherent: they describe crawling/aggregating tech and military news and list plausible sources. However, registry metadata (ownerId, slug, version) does not match the _meta.json and SKILL.md values, and homepage/source are missing — this mismatch suggests sloppy packaging or possible repackaging under a different identity.

Instruction Scope

Runtime instructions are limited to web searching/fetching (references to 'tavily' or 'web_fetch'), filtering duplicates/unreliable items, and producing structured summaries. The instructions do not request local file reads, secret access, or sending data to unexpected external endpoints beyond the listed news sites.

Install Mechanism

No install spec and no code files — instruction-only — so nothing is downloaded or written during install. This minimizes risk from malicious install mechanisms.

Credentials

The skill declares no required environment variables, binaries, or config paths. The operations it describes (web fetching public news sites) do not require credentials, so requested access is proportionate.

Persistence & Privilege

always is false and the skill is user-invocable; it does not request permanent presence or modifications to other skills. Autonomous invocation is allowed (platform default) but not combined with other privileges here.

What to consider before installing

This skill's behavior (fetching and summarizing news) matches its description and it is instruction-only, which limits risk. However, the package has inconsistent metadata (registry owner/slug/version differ from _meta.json and SKILL.md) and no published homepage or known source; that reduces provenance and increases risk of tampering or repackaging. Before installing: verify the publisher identity or prefer a skill with a known source, run it in a restricted environment, and avoid providing any credentials. If you plan to let the agent invoke the skill autonomously, be aware it will fetch arbitrary web pages — review the tool permissions (tavily/web_fetch) and content-handling policies to ensure fetched pages won't be forwarded to other external services.

News Aggregator

聚合国内外社会、科技、军事新闻，自动筛选要点。

新闻源

国内科技

• 36氪 (https://36kr.com/information/tech/)
• 机器之心 (https://www.jiqizhixin.com/)
• 量子位 (https://www.1baijia.com/)
• IT之家 (https://www.ithome.com/)

国内军事

• 观察者网 (https://www.guancha.cn/)
• 澎湃新闻 (https://www.thepaper.cn/)
• 腾讯军事 (https://new.qq.com/om/mil/)

国际科技

• TechCrunch
• The Verge
• Wired
• Ars Technica

国际军事

• Defense News
• Jane's Defence
• Military Times

工作流

1. 搜索 - 用 tavily 或 web_fetch 搜索各源
2. 筛选 - 过滤重复、过期、不可靠来源
3. 整理 - 按类别整理，每条含标题、来源、要点
4. 输出 - 生成结构化汇总

可信度规则

优先：

• 官方媒体报道
• 权威机构发布

谨慎：

• 论坛帖子
• 匿名消息
• 二手转载

输出格式

## 科技新闻

1. [标题](链接)
   来源：xxx | 时间：xxx
   要点：xxx

## 军事新闻

1. [标题](链接)
   来源：xxx | 时间：xxx
   要点：xxx