qinghu-shopee-cross-site
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFECREDENTIALS_UNSAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill instructions direct the agent to retrieve authentication credentials from environment variables
QINGHU_TOKENorQHKIT_TOKENfor API access, which is a standard method for managing secret keys. - [COMMAND_EXECUTION]: The agent is instructed to use network-enabled utilities like
curl,fetch, orrequeststo interact with the API endpointhttps://www.iqinghu.com/api/desktop/qh/mcp. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data from a third-party API, creating a vulnerability surface for indirect prompt injection.
- Ingestion points: Data describing brand distributions and category trends is fetched from the
iqinghu.comAPI into the agent's context as described inSKILL.md. - Boundary markers: There are no specific delimiters or prompts instructing the agent to ignore potentially malicious instructions within the retrieved API text.
- Capability inventory: The agent has network access capabilities and can invoke other skills like
qinghu-excel-exportto generate files. - Sanitization: The instructions do not specify any validation or sanitization steps for the text content received from the API payload.
Audit Metadata