linkfox-temu-manage-product-eu

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill contains numerous Python scripts designed to be executed by the agent to interact with Temu's Partner EU platform APIs. These scripts perform actions like querying goods, updating stock, and managing compliance information.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from Temu APIs and stores the results in the workspace. This data, which may include seller-controlled content such as product descriptions, represents a potential injection surface if processed directly by the agent.\n
  • Ingestion points: API responses from tool-gateway.linkfox.com are saved to local JSON files by the emit_result function in scripts/_temu_common.py.\n
  • Boundary markers: The skill instructions specifically advise the agent to use jq or ConvertFrom-Json to extract relevant fields, minimizing the exposure to potentially malicious content in the prompt context.\n
  • Capability inventory: The skill possesses the ability to execute shell commands and perform network requests.\n
  • Sanitization: The scripts do not perform explicit sanitization of the data retrieved from the API before storage.\n- [EXTERNAL_DOWNLOADS]: The onboarding.py utility references the installation of standard Python packages requests, qrcode, and pillow from the official PyPI registry to support account management and QR code generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 07:12 AM