skill-dropshipping-sourcing

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface through the retrieval of external content.
  • Ingestion points: Product metadata, including names and descriptions, is fetched from the external CJ Dropshipping API (/product/listV2) by the scripts/source.js script.
  • Boundary markers: The skill does not implement boundary markers, delimiters, or specific instructions to isolate fetched product content from system instructions in its output JSON files.
  • Capability inventory: The skill has the capability to write to the local file system using fs.writeFileSync in both scripts/source.js and scripts/token.js to store data and update authentication credentials.
  • Sanitization: No sanitization, filtering, or instruction detection is performed on the strings retrieved from the third-party API before they are written to the local system.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 09:34 PM