doubao-websearch

Pass

Audited by Gen Agent Trust Hub on Aug 11, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill sends queries to doubaoya.com to fetch search results. This is the primary function of the skill and targets the vendor's infrastructure.
  • [PROMPT_INJECTION]: By retrieving web content, the skill introduces a surface for indirect prompt injection. The risk is minimized by the skill's structure, which returns results in a specific JSON format for the agent to parse.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 11, 2026, 07:48 PM
Security Audit — agent-trust-hub — doubao-websearch