bing-search

Pass

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill transmits search parameters and an authentication token to the API at ai-factory.frevana.com. This communication is essential for the skill's primary function of retrieving search data from the author's service.
  • [COMMAND_EXECUTION]: Local shell and Python scripts are used to manage input parameters and ensure the API response is valid JSON. All processing occurs within the skill's local environment.
  • [PROMPT_INJECTION]: As a data-retrieval tool, the skill processes information from external web results, which creates a potential surface for indirect prompt injection.
  • Ingestion points: scripts/search_bing.sh (retrieves results from the API).
  • Boundary markers: None present.
  • Capability inventory: curl, bash, python3, and file writing via --output.
  • Sanitization: None; returns raw results.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 26, 2026, 08:37 AM
Security Audit — agent-trust-hub — bing-search