qinghu-1688-goods-detail

Pass

Audited by Gen Agent Trust Hub on Nov 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to download the Node.js runtime environment from nodejs.org and the qhkit package from official npm registries.
  • [REMOTE_CODE_EXECUTION]: Outlines a secure installation process for the Node.js runtime, featuring SHA256 integrity verification of the downloaded binary. While automated scanners flagged a command chain as piped to shell, the code specifically pipes to grep and sha256sum for validation purposes, following security best practices.
  • [COMMAND_EXECUTION]: Executes shell commands for environment setup and tool operations, providing specific mitigations for Windows PowerShell character encoding and JSON parsing issues.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied 1688 offer IDs and links for querying the API. It mitigates potential command injection risks by recommending the use of a JSON parameter file (@params.json) to pass arguments to the qhkit CLI safely.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 7, 2026, 01:42 AM