linkfox-amazon-store-uploads

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local dependency check script using the Python interpreter. The script scripts/_spapi_uploads_common.py calls subprocess.run to verify the presence of the linkfox-amazon-store-auth skill by running scripts/check_auth_dependency.py.
  • [DATA_EXFILTRATION]: The skill reads local files and transmits them to external endpoints.
  • Ingestion: File paths are provided via JSON parameters in filePath.
  • Transmission: Files are uploaded to URLs provided by the Amazon SP-API or via the vendor's developer proxy.
  • Vendor Infrastructure: All network operations target linkfox-ai vendor domains (tool-gateway.linkfox.com, api.linkfox.com, agent-api.linkfox.com).
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect injection as it processes externally-controllable file paths and URLs without strict validation.
  • Ingestion Points: sys.argv input in create_upload_destination_for_resource.py and upload_to_destination.py.
  • Boundary Markers: Absent; the skill does not wrap inputs in delimiters to prevent the agent from misinterpreting path strings as instructions.
  • Capability Inventory: File system read access (Path.read_bytes) and network write capabilities (urllib.request.urlopen with PUT/POST).
  • Sanitization: None; input paths are processed using standard path expansion (expanduser).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 07:11 AM