news
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a custom CLI tool
cyeamto retrieve structured news data based on a user-provided date. This is the primary function of the skill. - [EXTERNAL_DOWNLOADS]: The skill instructions specify 'secondary crawling' of news article links to fetch full text for AI summarization. This involves fetching content from external, third-party URLs.
- [INDIRECT_PROMPT_INJECTION]: Because the skill crawls and processes content from external news links, it is exposed to potential indirect prompt injection if an external article contains malicious instructions intended to hijack the agent's behavior.
- Ingestion points: External URLs found in the
linkfield of retrieved news items. - Boundary markers: None explicitly defined in the instructions.
- Capability inventory: Subprocess execution via
cyeamCLI and text summarization capabilities. - Sanitization: The skill relies on the agent's default processing; no specific sanitization of external content is mentioned.
Audit Metadata