toutiao-news-trends
Pass
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. It retrieves news titles and metadata from a third-party source (Toutiao) and presents them to the AI agent. If the external content contains maliciously crafted instructions, it could potentially influence the agent's behavior.
- Ingestion points: The
scripts/toutiao.jsfile fetches news data fromhttps://www.toutiao.com/hot-event/hot-board/. - Boundary markers: The skill does not use specific delimiters or instructions to the agent to ignore potential commands within the news data.
- Capability inventory: The skill itself does not possess dangerous capabilities like file writing, network exfiltration of local data, or arbitrary command execution beyond running its own script.
- Sanitization: No sanitization or filtering of the fetched news titles or labels is performed prior to output.
Audit Metadata