Security audit
Security checks for vulnerabilities and agentic risk
This skill is a local meeting follow-up formatter that generates reviewable text and does not send email, access networks, install dependencies, or persist changes beyond an optional user-specified output file.
Install only if you are comfortable with a Chinese-language local helper that reads meeting notes you provide and optionally writes a local report file. It does not send email for you; review generated drafts before using them. The package contains unused audit helper code, so publishers should document or remove that code before exposing any audit mode.
声明描述的是一个会议后续事项整理与邮件草稿辅助工具；而实际代码实现的是一个由 spec 驱动的通用本地分析/审计 CLI。其核心能力包括扫描目录、统计文件类型、读取 CSV、做模式匹配、检查 Skill 目录规范等，这些都与“会后事项拆解、催办节奏、邮件草稿、下次同步议题”没有直接对应关系。虽然 make_structured_report 可把输入文本按 section 组织成结构化 Markdown，看起来可被某些 spec 配置用于摘要整理，但代码的整体主用途明显更广且偏审计/分析，不是所声明的会议 follow-up 专用能力。因此属于描述与实际行为的实质性不匹配。
The script implements broad directory inspection, CSV auditing, pattern scanning, and skill-package auditing behaviors that materially exceed the declared purpose of a follow-up/meeting workflow skill. That mismatch is dangerous because it enables collection and summarization of unrelated local repository content, increasing the chance of unintended data exposure and making the skill a stealthy reconnaissance tool.
The README uses Chinese throughout, including the title, usage guidance, examples, and risk notices, but does not indicate that the skill is region-specific or offer users an alternative language. Under the policy rule, forcing a specific language without opt-in is a natural-language policy violation.
This markdown file presents all user-facing audit content in a single language, which can violate language/locale policy when no user opt-in or locale justification is provided. The file does not indicate that the skill is China-specific or that Chinese output is optional.
The skill advertises executable capabilities via python3 and references local input/output files and resources, but it does not declare any explicit tool scope such as allowed tools or permissions. This creates an authorization ambiguity where a seemingly simple follow-up drafting skill may gain shell and file access beyond what a reviewer or user would reasonably expect.
The natural-language content of the skill is written as mandatory Chinese-facing instructions and output structure, with no indication that users may choose another language. This creates a language/locale policy issue because the skill appears to constrain interaction language without documented opt-in or justification.
The entire example input is written in Chinese and provides no indication that users may supply or receive content in other languages. Under the policy criteria, a language-specific constraint without user opt-in or a documented justification can constitute a natural-language policy violation.
This markdown file presents all headings and instructions in Chinese, and nowhere indicates that the user can choose another language or that the skill is intentionally limited to Chinese-language output. Under the language/locale policy, forcing a specific language without user opt-in is a natural-language policy violation.
The template headings and instructions are entirely in Chinese, which imposes a specific language on outputs. Under the policy rule, locale or language constraints should either offer user opt-in/choice or be clearly documented as justified for a region-specific use case, which is not present here.
The code can recursively enumerate files and read contents from directories, then extract headings and scan for secret-like strings and private URLs. In the context of a meeting follow-up skill, those capabilities are unjustified and create unnecessary access to potentially sensitive local data, even if the current implementation only reports excerpts and matches.
This code generates multiple user-facing report headings and guidance strings exclusively in Chinese, such as "结果", "摘要", and "下一步". Because the script does not offer a language option or document that it is intentionally region-specific, it violates the language/locale policy for natural-language behavior.
The directory report mode emits headings and guidance entirely in Chinese, including "扫描报告", "目录概览", and review instructions. This is another instance of forcing a specific language on users without an explicit choice or clear regional constraint.
Several functions generate user-visible content only in Chinese, including CSV summaries, pattern scan results, and skill audit messages. Across these modes, the skill consistently forces one language instead of offering an opt-in or documenting a justified locale restriction.
The skill-audit and frontmatter parsing logic inspects package structure and metadata unrelated to follow-up planning. That broadens the skill into a generic auditing tool, which can be abused to inventory local project contents and violates least-privilege expectations for users invoking a meeting/email assistant.
This markdown file is entirely written in Chinese, including the title, test steps, and pass criteria, with no indication that the user can choose another language. Under the stated policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale restriction is justified.
No suspicious patterns detected.
