Security audit
Security checks for vulnerabilities and agentic risk
This skill fetches public trending-topic data from disclosed websites and does not show hidden persistence, credential access, local data collection, or destructive behavior.
Install this if you want a Chinese-oriented trend scraper for public Baidu, Toutiao, and GitHub Trending data. Use it intentionally because it makes outbound requests to those sites, and consider narrowing its trigger phrases if your agent auto-loads skills aggressively.
The skill explicitly advertises live scraping from external sources and the usage examples invoke Python code that performs network access, but the manifest does not declare any tool scope such as permissions or allowed-tools. That mismatch weakens security boundaries and reviewability because an agent or platform cannot clearly constrain or warn about external data access before the skill runs.
The description includes broad trigger phrases like 热搜, 热榜, 趋势, trending, and 挖掘需求, which are common in ordinary conversation and can cause the skill to auto-match in contexts the user did not intend. Over-broad invocation increases the chance of unsolicited network fetching and external content ingestion, which expands prompt-injection and privacy exposure from untrusted web data.
The module docstring presents the skill description only in Chinese ("真实热搜抓取"). This imposes a fixed language/locale in natural-language content without offering a user choice or documenting a justified region-specific constraint.
The user-facing natural-language description and instructions are entirely in Chinese, with no indication that users may choose another language or locale. Under the stated policy, language constraints should be optional or clearly justified; no such opt-in or justification is present here.
This code fetches data from multiple external services using urllib.request.urlopen, but the file provides no explicit user-facing notice about making network requests. While network access is central to the script's purpose, there is no confirmation, comment, or disclosure that queries will be sent to Baidu, Toutiao, and GitHub.
No suspicious patterns detected.
