toutiao-news-trends
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes news titles and labels from an external source, creating a surface for potential indirect prompt injection attacks.
- Ingestion points: The
scripts/toutiao.jsscript fetches data fromhttps://www.toutiao.com/hot-event/hot-board/. - Boundary markers: The output uses
---delimiters and Markdown headers to separate news items, but lacks explicit "ignore embedded instructions" warnings for the agent processing the output. - Capability inventory: The skill has network read access and prints to stdout. It does not have file-write or shell-execution capabilities beyond running its own script.
- Sanitization: The script cleans query and hash parameters from detail links but does not sanitize the text content of the news titles or labels.
- [COMMAND_EXECUTION]: The skill relies on the execution of a local Node.js script (
scripts/toutiao.js) via the command line to perform its data fetching and formatting tasks. The command logic is fixed and does not allow for arbitrary shell injection from user inputs.
Audit Metadata