perplexity-research

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The client script makes outbound HTTPS connections to api.aisa.one. This domain belongs to the skill's vendor (aisa-team) and is used exclusively for fetching research and search data. This represents normal vendor functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill serves as a conduit for web content, which could contain malicious instructions. The script retrieves data from search results and extracts content from user-provided URLs.
  • Ingestion points: scripts/search_client.py receives data from the AIsa API endpoints (e.g., /sonar, /tavily/extract).
  • Boundary markers: The script does not implement explicit boundary markers for the retrieved content.
  • Capability inventory: Network access is limited to the vendor's API domain.
  • Sanitization: The script displays snippets and answers as returned by the API service.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 02:35 AM