toutiao-news-trends
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill uses standard Node.js built-in modules (
https,zlib) to perform its stated function. All network communications are directed towww.toutiao.com, which is a well-known news service. No evidence of credential theft, remote code execution, or persistence was found. - [INDIRECT_PROMPT_INJECTION]: The skill ingests news headlines and metadata from an external website, which constitutes an attack surface where malicious content could be used to influence the agent's behavior.
- Ingestion points: Data is fetched from the public Toutiao API at
https://www.toutiao.com/hot-event/hot-board/withinscripts/toutiao.js. - Boundary markers: None; the skill does not explicitly instruct the agent to treat the retrieved content as potentially untrusted.
- Capability inventory: The skill is restricted to fetching and outputting JSON data to the console. It does not perform file system modifications, arbitrary command execution, or exfiltration of local user data.
- Sanitization: The script sanitizes links by removing query parameters but does not filter or sanitize the news headlines or category labels before output.
Audit Metadata