Ai News Hub
AdvisoryAudited by VirusTotal on Apr 14, 2026.
Overview
Type: OpenClaw Skill Name: ai-news-hub Version: 2.3.1 The ai-news-hub skill bundle is a legitimate tool for aggregating AI-related news, research papers, and GitHub projects. The Python scripts (rss_aggregator.py, arxiv_papers.py, github_trending.py, and summarize_url.py) use standard libraries to fetch public data from well-known sources like arXiv, GitHub, and various tech blogs. The code implements responsible practices such as concurrency, caching (ETag/Last-Modified), and timeouts. No evidence of malicious intent, data exfiltration, or prompt injection was found; the functionality is entirely consistent with the stated purpose of a news aggregator.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If used on private, internal, or token-bearing URLs, the skill could retrieve content the user did not intend to process through the agent.
The article summarizer accepts and fetches an arbitrary user-provided URL. This is useful for summarizing public articles, but it is broader than the fixed RSS source list.
parser.add_argument("url", help="要抓取的文章 URL") ... html = fetch_article(args.url)Use the URL summarizer only for public news/article links, and consider adding host allowlists or confirmation before fetching non-public URLs.
Public article URLs are fine for this workflow, but private URLs, signed links, or URLs containing sensitive query parameters could be exposed to the third-party reader service.
When direct extraction is insufficient, the script can send the requested URL to Jina Reader, an external third-party reader service.
jina_url = f"https://r.jina.ai/{url}" ... print("直接抓取内容不足，尝试 Jina Reader...", file=sys.stderr)Avoid using private or authenticated URLs with this feature, or make third-party reader use opt-in and clearly disclosed.
Cached public feed content can improve performance, but stale or malicious feed text could be reused in later news results if a source is compromised.
Fetched RSS content and metadata are cached locally for reuse for about one hour.
CACHE_PATH = Path(__file__).parent / ".rss_cache.json" ... "content": content
Treat RSS/article text as untrusted content, do not follow instructions embedded in retrieved articles, and clear the cache if results look suspicious or outdated.
Manual installs from remote sources carry ordinary package/provenance risk if the repository or package changes.
The manual installation documentation references cloning from GitHub and optionally installing an unpinned Python package. This is user-directed and not automatically executed by the skill.
git clone https://github.com/lanyasheng/ai-news-aggregator.git ... pip3 install feedparser
Install only from trusted sources, consider pinning package versions, and review repository changes before manual installation.