ebay-search

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local bash script (scripts/search_ebay.sh) which uses curl for network requests and python3 for JSON payload construction and validation.
  • [EXTERNAL_DOWNLOADS]: The skill connects to https://ai-factory.frevana.com to fetch eBay listing data as part of its primary function.
  • [DATA_EXFILTRATION]: The skill transmits search queries and an authentication token (FREVANA_TOKEN) to the vendor's API endpoint. This is standard behavior for an API-based search tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data from eBay listings, which represents an attack surface for indirect prompt injection.
  • Ingestion points: The API response is downloaded and processed by the agent to provide summaries to the user.
  • Boundary markers: There are no explicit delimiters or instruction-ignore warnings applied to the external data before it is presented to the agent.
  • Capability inventory: The skill possesses network access (curl) and file system write access (cp, mkdir).
  • Sanitization: The script includes a Python-based validation step to ensure that the API response is valid JSON before it is saved or returned.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 12:23 AM
Security Audit — agent-trust-hub — ebay-search