search-specialist

Pass

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: LOWPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • Indirect Prompt Injection (LOW): The skill is designed to fetch and synthesize content from external websites, which is a known attack surface for indirect prompt injection.
  • Ingestion points: Web search results and full content extraction via WebFetch.
  • Boundary markers: Not present in the provided instructions.
  • Capability inventory: Data synthesis and recommendation generation; no file-write, network exfiltration of sensitive local data, or command execution capabilities are present.
  • Sanitization: No sanitization or filtering logic is defined for the external content.
  • Data Exposure & Exfiltration (LOW): The skill performs network operations to fetch data from non-whitelisted external domains as part of its core search and 'WebFetch' functionality.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 15, 2026, 11:25 PM