Security audit
Security checks for vulnerabilities and agentic risk
This skill is a local, user-invoked deadline-summary helper with no network access or automatic system changes, though its bundled script contains unused generic audit helpers that should be cleaned up or documented.
Install only if you want a Chinese-language local helper for reviewing deadline and schedule information. Provide specific task, calendar, or message-summary files rather than broad folders or sensitive exports, and treat the output as a review draft, not an automatic calendar update.
声明描述的是个人生产力场景中的待办/会议/消息摘要分析，重点应是发现截止日期和日程冲突。但代码并未接入日历、待办或消息对象，也没有任何与时间解析、截止日期识别、冲突检测相关的逻辑。相反，它读取 spec.json 决定模式，执行目录扫描、CSV 概览、正则模式匹配、Skill 结构审计，并输出通用报告。这属于 materially different primary purpose，并且实现了多项未声明的审计能力。虽然代码本身主要是只读分析、未发现自动修改日历之类行为，但其实际用途与声明严重不符，因此应判定为 mismatch。
The pattern scanning logic searches files for secrets, internal URLs, and dangerous shell snippets, which is unrelated to the declared scheduling workflow. In context, this is dangerous because it enables the skill to inspect arbitrary content for sensitive security artifacts, suggesting broader reconnaissance behavior than users would reasonably expect from a deadline assistant.
The dispatch logic implements a generic auditing framework for directories, CSV files, arbitrary text, pattern scans, and skill-package checks rather than a narrowly scoped personal deadline analyzer. In a skill that claims to process deadlines, meetings, and message summaries, this creates a capability mismatch that can be used to inspect unrelated local content, increasing the risk of over-collection and unintended disclosure of sensitive data.
The trigger phrases are short and generic, such as asking to find deadline risks or identify time conflicts, without explicit scoping to personal planning inputs. In a skill-routing system, this can cause unintended activation on loosely related prompts and may route sensitive calendar, message, or task content into this skill when the user did not clearly intend it.
The skill advertises executable shell usage and implies file input/output behavior, but it does not declare an explicit permission or allowed-tools scope. That creates an authorization gap: a host may expose broader capabilities than intended, and users or orchestrators cannot reliably tell what the skill is allowed to access or modify.
This markdown example output forces a specific language/locale for all headings and user-facing text. Under the policy, language constraints should either be user-selectable or clearly justified as region-specific; neither is present in this file.
The title, category label, summary, and input hint are written in Chinese, which indicates the skill is designed to operate in a specific language/locale. Under the policy, locale constraints should either be optional for the user or clearly documented as justified; this manifest does not provide opt-in language choice or a region-specific rationale.
The recursive file listing function walks arbitrary directories and selects many text-like file types for inspection. For a personal deadline radar skill, this broad filesystem crawling is unjustified and increases the chance of reading sensitive notes, code, configuration, and exported data that are unrelated to the advertised task.
The generated user-facing report strings are written entirely in Chinese, including headings, summaries, and action prompts. Because this code file does not provide a user opt-in or configurable locale, it appears to force a specific language, which matches the language/locale policy violation criteria.
The directory and CSV reporting functions emit fixed Chinese headings and guidance text to the user. This repeats the same forced-locale behavior across multiple code paths, with no opt-in, fallback, or documented region-specific justification.
The CLI framing presents the program as a routine support script for the skill, while the actual behavior is a generic report generator with multiple unrelated audit modes. This mismatch can mislead operators about what data will be processed, undermining informed consent and making potentially invasive analysis appear harmless.
This markdown file presents all user-facing instructions exclusively in Chinese. Under the policy for natural-language violations, forcing a specific language without user opt-in can be a policy issue when no alternative or choice is documented.
The README presents the skill name, instructions, examples, and warnings entirely in Chinese, which may implicitly force a specific language experience. There is no statement that the user may choose another language or that the Chinese-only presentation is intentional for a region-specific audience.
This markdown file presents all user-facing content in Chinese only. Under the language/locale policy rule, forcing a specific language without offering user choice or documenting a justified locale constraint is a natural-language policy concern.
The manifest description and routing examples are written in a mixed Chinese/English style, and the skill name/body otherwise present the skill as operating in Chinese without stating that users may choose their preferred language. This can amount to an implicit language constraint or inconsistent locale behavior without explicit opt-in.
The file's natural-language instructions and labels are all in Chinese, and there is no indication that users may opt into another language or that the skill is intentionally limited to a Chinese-language audience. This can violate a language/locale policy when a skill implicitly forces a specific language without user choice.
The entire template is written in Chinese and presents section headings and instructions only in Chinese, which implies a fixed language requirement. Under the policy rule, forcing a specific language without offering user choice or documenting a justified locale constraint is a natural-language policy violation.
No suspicious patterns detected.
