Security audit
Security checks for vulnerabilities and agentic risk
This is a prompt-only translation skill with disclosed optional file or Feishu document output, so it is generally safe but users should be careful before sending sensitive text to external documents.
Install this as a normal translator skill, but only ask it to write files or create Feishu documents when you intend that content to be saved there; avoid using external document output for confidential text unless that destination is approved.
The instruction defaults output to Chinese or English when the user does not specify a target language, which imposes a language choice automatically. This is a locale/language policy concern because the skill does not ask for user preference or offer an opt-in before selecting the output language.
The trigger "Any request involving converting text from one language to another" is extremely broad and lacks scope limits or exclusion conditions. This could cause the skill to activate on a wide range of ordinary conversation or adjacent editing tasks without clear boundaries.
The skill is presented as a zero-dependency, pure prompt-driven translator, but its documented delivery options include writing files and creating/writing Feishu documents. This creates a capability mismatch that can mislead operators and users about side effects, increasing the risk of unintended external writes or data handling beyond expected translation-only behavior.
Including Feishu document creation/writing introduces an external data exfiltration and persistence channel that is not central to a basic translator skill. If invoked on sensitive source text, the skill could send confidential content to a third-party document system, which materially expands risk compared with an in-chat translation-only workflow.
No suspicious patterns detected.
