linkfox-temu-manage-product-global
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting and processing external product data from Temu APIs. This data (e.g., goods names, descriptions) is written to local files and displayed to the agent without sanitization, which could allow a malicious actor controlling the platform data to influence the agent's behavior.
- Ingestion points:
scripts/_temu_common.pyprocesses responses from/temu/proxyand/temu/fileDownload. - Boundary markers: No explicit delimiters or 'ignore instructions' warnings are used when interpolating API data into the agent's context.
- Capability inventory: The skill possesses file-write capabilities and network access to the vendor's gateway.
- Sanitization: Data is parsed as JSON, but individual string fields are not escaped or validated for malicious instructions.- [DATA_EXFILTRATION]: The skill automatically writes detailed API responses to a local
linkfoxdirectory in the current working directory or home folder via theemit_resultfunction inscripts/_temu_common.py. This behavior exposes potentially sensitive business data to the local filesystem.
Audit Metadata