qinghu-ecom-sourcing
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill instructions direct the agent to retrieve authentication tokens from environment variables (
QINGHU_TOKENorQHKIT_TOKEN) and transmit them to the vendor's API endpoint athttps://www.iqinghu.com. While this is standard for authenticated API access, it involves the retrieval and external transmission of sensitive environment data to a non-whitelisted domain.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple external platforms, creating a potential attack surface for indirect prompt injection.\n - Ingestion points: Data retrieved from external marketplaces (Amazon, TikTok, Ozon, 1688, etc.) via the
iqinghu.comAPI, specifically through thetools/callmethod.\n - Boundary markers: No specific delimiters, framing, or instructions to ignore embedded commands are defined for the data processed from the API responses.\n
- Capability inventory: The skill uses network operations to fetch data and includes capabilities for file export (via the
qinghu-excel-exportskill) and local file caching as described in the workflow.\n - Sanitization: The instructions do not specify any validation or sanitization routines for the data ingested from external sources before it is used to generate reports, perform further analysis, or be exported to files.
Audit Metadata