news
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill relies on fetching external data from third-party news websites, creating an entry point for untrusted instructions that could influence the agent's behavior.
- Ingestion points: News site URLs listed in
SKILL.mdaccessed dynamically viabrowser_usetool actions (openandsnapshot). - Boundary markers: Absent. The skill does not instruct the agent to enclose the incoming web content in distinct delimiters or explicitly ignore embedded commands within the text.
- Capability inventory: None. No custom tools or system commands are registered within this skill.
- Sanitization: Absent. The content fetched from the webpage snapshots is processed and presented directly without validation or filtering.
Audit Metadata