linkfox-amazon-store-fba

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill performs network operations to interact with Amazon SP-API via the vendor's infrastructure.
  • It communicates with tool-gateway.linkfox.com, api.linkfox.com, and agent-api.linkfox.com for API proxying and account management functions in _spapi_fba_common.py and onboarding.py.
  • These domains are verified vendor resources belonging to linkfox-ai and represent standard functionality for the skill's purpose.
  • [COMMAND_EXECUTION]: The skill executes a local script to verify environment dependencies.
  • In _spapi_fba_common.py, subprocess.run is used to execute check_auth_dependency.py using the current Python interpreter (sys.executable). This is a benign internal check to ensure the required authentication skill is installed.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data returned from external Amazon SP-API calls, which serves as an ingestion point for untrusted data.
  • Ingestion points: API responses processed in _fba_runner.py and _spapi_fba_common.py (specifically the payload field).
  • Boundary markers: Not present; the skill relies on the structured JSON nature of the API responses.
  • Capability inventory: The skill has capabilities to write data to the local filesystem (in the linkfox/ directory) and perform network requests to the vendor gateway.
  • Sanitization: Standard JSON parsing is performed, but no specific prompt-level sanitization is applied to the ingested data before it is presented to the agent. The risk is considered low given the highly structured nature of FBA business data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 07:11 AM