skills/unbrowse-ai/sites/wikivoyage/Gen Agent Trust Hub

wikivoyage

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides integration with Wikivoyage through the Unbrowse platform, a vendor-owned service. All remote connections are directed to the vendor's official domain at unbrowse.ai.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data from wikivoyage.org, creating a potential surface for indirect instructions.
  • Ingestion points: The en_wikivoyage_org__render_page tool in SKILL.md fetches content from Wikivoyage.
  • Boundary markers: No explicit delimiters are used to wrap external content in the provided instructions.
  • Capability inventory: The skill relies on remote MCP tools for data retrieval and task execution.
  • Sanitization: Results are obtained through verified first-party API replays rather than raw web scraping.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 04:13 AM