Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only helper for Word mail merge and DOCX template workflows, with no evidence of hidden execution, data theft, or persistence.
Use this skill for Word mail merge or DOCX template automation only. Review generated documents carefully before using them in legal, HR, sales, or administrative workflows, especially because the published trigger wording is broader and less polished than the actual skill instructions.
The trigger sentences are malformed and extremely broad, including generic phrases like 'Help me' and 'I need a practical workflow' followed by requirement text. This can cause accidental invocation on ordinary user requests that are not specifically about this skill, leading to prompt-routing confusion, over-broad activation, and possible misuse of the skill in unintended contexts.
The trigger sentences are generic and malformed enough that they could match unrelated user requests and cause unintended activation of this skill. In an agent environment, overly broad routing increases the chance of inappropriate tool/skill selection, which can mis-handle user data or produce incorrect workflow actions even though the content here is not overtly malicious.
The trigger phrases are broad and loosely scoped, which can cause the skill to activate for generic Word or workflow requests that were not intended for this specialized helper. In an agent environment, overbroad activation increases the chance of misrouting tasks, applying the wrong workflow, or injecting unnecessary instructions into unrelated user sessions.
The activation description is broad enough to match many generic productivity or document-automation requests, which can cause unintended skill invocation outside its intended scope. In an agent setting, overbroad routing can misapply specialized instructions or outputs to unrelated tasks, reducing reliability and potentially bypassing more appropriate skills.
Including a broad keyword like "work-productivity" creates a catch-all trigger that may activate on many unrelated user requests. This increases the chance of incorrect skill selection, which is dangerous in multi-skill agents because it can steer execution, instructions, or generated artifacts down an unintended path.
The skill enables implicit invocation without any trigger constraints, allowing the agent to call it automatically based only on broad relevance. Because this skill can influence document-generation workflows for legal, HR, sales, and administrative use cases, unintended invocation could cause prompt-scope expansion, misuse in sensitive business contexts, or execution of adversarial skill instructions without explicit user intent.
The README states that the skill provides English and Chinese instruction and guide files, but it does not explain how the user's preferred language is selected. This can create a locale-policy concern if the skill defaults to one language without explicit user choice.
This is a Chinese-language README, but the listed trigger sentences are only provided in English, suggesting the user may need to invoke the skill in English rather than being offered a language choice. That can conflict with language/locale policy expectations when no opt-in or justification is provided.
The file is presented as a zh-CN skill, but the activation keywords and example trigger sentences are written in English and presented as the expected invocation form. This may impose an English-centric interaction pattern without explicit user opt-in or a documented language choice.
No suspicious patterns detected.
