dreamina-inspiration-search

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • Prompt Injection (LOW): The skill is susceptible to indirect prompt injection (Category 8) because it retrieves and processes content (titles and AI prompts) from an external community database.
  • Ingestion points: Data enters via the item_list array in the JSON response from jimeng.jianying.com.
  • Boundary markers: There are no explicit instructions or delimiters to prevent the agent from obeying instructions embedded within the retrieved prompts.
  • Capability inventory: The skill uses the requests library for network operations to communicate with the service API.
  • Sanitization: The skill performs no filtering or sanitization of the prompt or title fields returned by the API before they are presented to the agent context.
  • Data Exposure & Exfiltration (SAFE): The skill requires a sessionid for authentication. However, it does not hardcode credentials or access local sensitive files; the token is passed as a function parameter, which is standard for authenticated API tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:36 PM