news
Pass
Audited by Gen Agent Trust Hub on Mar 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses established news organizations as information sources. No attempts at prompt injection, obfuscation, or data exfiltration were found.- [SAFE]: Tool usage is restricted to the
browser_usetool for navigation and snapshotting of specific URLs, posing no risk of unauthorized command execution.- [SAFE]: The skill has an indirect prompt injection surface through external data ingestion. 1. Ingestion points: Specified news URLs. 2. Boundary markers: None. 3. Capability inventory: Browser snapshot and navigation. 4. Sanitization: None. This surface is assessed as safe because the URLs are authoritative and the agent's actions are limited to summarization.
Audit Metadata