低风险 — 风险评分 15/100
上次扫描:17 小时前 重新扫描
15 /100
ai-agent-news-aggregator
AI Agent 资讯聚合技能 - 搜集 AI Agent 领域最新资讯并推送到飞书群聊
A legitimate news aggregation skill with minor doc-to-code inconsistencies but no malicious behavior detected.
技能名称ai-agent-news-aggregator
分析耗时39.9s
引擎pi
可以安装
This skill is safe to use. Consider documenting the subprocess usage in SKILL.md for transparency, but no immediate security action is required.

安全发现 2 项

严重性 安全发现 位置
低危
Undeclared subprocess execution 文档欺骗
run_pipeline.py uses subprocess.run() to execute Python scripts in the scripts/ directory, but SKILL.md does not declare shell execution as a capability.
result = subprocess.run(cmd, capture_output=True, text=True)
→ Add shell:WRITE to the declared capabilities in SKILL.md if subprocess execution is intended
scripts/run_pipeline.py:19
提示
Tool name mismatch 文档欺骗
SKILL.md mentions 'ddg-search' and 'blogwatcher' as external dependencies, but the code only constructs URLs without actually invoking these tools.
SKILL.md: 'ddg-search' - DuckDuckGo 网页搜索
→ Update SKILL.md to reflect that URLs are built and returned for external fetching
SKILL.md vs scripts/search_news.py:36
资源类型声明权限推断权限状态证据
文件系统 NONE READ ✓ 一致 Reads config JSON files for keywords and channel_id
网络访问 READ READ ✓ 一致 Makes HTTP requests to DuckDuckGo and RSS feeds for news aggregation
命令执行 NONE WRITE ✓ 一致 run_pipeline.py:19 uses subprocess.run() to execute Python scripts
8 项发现
🔗
中危 外部 URL 外部 URL
https://news.ycombinator.com/newest
SKILL.md:202
🔗
中危 外部 URL 外部 URL
https://www.anthropic.com/news/rss.xml
SKILL.md:203
🔗
中危 外部 URL 外部 URL
https://.../rss.xml
scripts/README.md:84
🔗
中危 外部 URL 外部 URL
https://lite.duckduckgo.com/lite/
scripts/search_news.py:32
🔗
中危 外部 URL 外部 URL
https://hnrss.org/newest?q=ai+agent
scripts/sources.json:15
🔗
中危 外部 URL 外部 URL
https://huggingface.co/blog/feed.xml
scripts/sources.json:27
🔗
中危 外部 URL 外部 URL
https://blog.langchain.dev/rss/
scripts/sources.json:31
🔗
中危 外部 URL 外部 URL
https://lite.duckduckgo.com/lite/?q=AI+Agent+framework+OR+LangChain+new+release+OR+AutoGen+update+OR+CrewAI+OR+LlamaInde...
scripts/step1_search.json:2

目录结构

12 文件 · 33.9 KB · 1367 行
Python 7f · 893L Markdown 2f · 400L JSON 2f · 64L Text 1f · 10L
├─ 📁 scripts
│ ├─ 🐍 categorize.py Python 139L · 3.9 KB
│ ├─ 🐍 deduplicate.py Python 145L · 3.8 KB
│ ├─ 🐍 push_to_feishu.py Python 168L · 4.8 KB
│ ├─ 📝 README.md Markdown 137L · 2.3 KB
│ ├─ 📄 requirements.txt Text 10L · 351 B
│ ├─ 🐍 run_pipeline.py Python 145L · 4.6 KB
│ ├─ 🐍 search_news.py Python 111L · 3.2 KB
│ ├─ 📋 sources.json JSON 49L · 1.2 KB
│ ├─ 📋 step1_search.json JSON 15L · 457 B
│ ├─ 🐍 summarize.py Python 108L · 2.9 KB
│ └─ 🐍 test_push.py Python 77L · 1.8 KB
└─ 📝 SKILL.md Markdown 263L · 4.6 KB

依赖分析 1 项

包名版本来源已知漏洞备注
none (standard library only) N/A pip No external packages installed

安全亮点

✓ No external dependencies required (uses Python standard library only)
✓ No credential harvesting or sensitive path access
✓ No base64 encoding, obfuscation, or anti-analysis techniques
✓ No network exfiltration or C2 communication
✓ No remote script execution (curl|bash, wget|sh)
✓ No hidden functionality - all scripts perform declared tasks
✓ No supply chain risks (no unpinned dependencies)
✓ Legitimate news aggregation use case with transparent behavior