agency-corporate-training-designer
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were detected. The skill consists entirely of behavioral instructions and domain knowledge for a 'Corporate Training Designer' persona.
- [INDIRECT_PROMPT_INJECTION]: The skill describes workflows for processing untrusted external data, such as employee surveys, performance reviews, and 360-degree feedback. However, it does not provide any scripts or tools that could be exploited via this data. The risk is mitigated by the lack of execution capabilities within the skill's own code.
- [EXTERNAL_DOWNLOADS]: The instructions reference several well-known Chinese enterprise learning platforms (e.g., DingTalk, WeCom, Feishu, UMU). These are mentioned as part of the domain expertise for the persona and do not involve automated network operations, API interactions, or software downloads.
Audit Metadata