Security audit
Security checks for vulnerabilities and agentic risk
This skill is a simple document text extractor with some inaccurate read/write wording, but I found no hidden, destructive, persistent, or exfiltrating behavior.
Install only if you want a Chinese-documented helper for extracting text from DOCX, PDF, and Excel files. Do not rely on it for editing documents; the advertised write/edit paths are not implemented.
该代码的核心功能是按文件扩展名读取 Word、PDF 和 Excel 内容并输出文本，和“读取”部分基本一致。但声明明确写了“读取和编辑”，而实际没有任何修改、保存、写入文档的逻辑，因此存在能力夸大。另一个轻微偏差是声明提到 openpyxl，但代码中 Excel 处理实际依赖 pandas（虽然 pandas 可能间接使用相关引擎，但从代码表面并非所述实现）。总体上属于描述未准确反映实际行为，应判定为不匹配。
The manifest description, trigger phrases, and headings are all written only in Chinese, which imposes a language choice without indicating user opt-in or that the skill is intentionally limited to a Chinese-speaking context. This can conflict with language or locale policy expectations when no explicit language-selection behavior is documented.
The read_when entries such as '读取文档' and '编辑文档' are generic descriptions that could overlap with many ordinary document-related requests. The file does not provide narrower trigger conditions, exclusions, or negative examples to clarify when this skill should or should not activate.
Natural-language strings in the docstring and CLI output are written entirely in Chinese, including usage, error, and unsupported-command messages. For a general document-processing tool, this imposes a specific language/locale on users without opt-in or a documented region-specific justification, which matches the language-policy violation criteria.
The manifest description says the skill can '读取和编辑 Word、PDF、Excel 文档', which implies both reading and editing. However, the code only parses and returns content from DOCX, PDF, and Excel files and exposes only a 'read' command path in the CLI.
No suspicious patterns detected.
