Security audit
Security checks for vulnerabilities and agentic risk
This skill is a local Chinese-language escalation brief writer with no evidence of network access, persistence, destructive actions, or hidden automatic execution.
Install only if you want a Chinese-language local helper for drafting escalation/support briefs. Provide only the files or text you want summarized, and be aware the bundled script contains unused audit capabilities that should remain disabled unless intentionally reviewed and documented.
声明描述的是一个面向升级说明/支持沟通的整理工具，核心应是把输入信息重组为背景、影响、已尝试、诉求等结构化简报。代码虽然在 mode 为 structured_brief 时确实支持这类功能，但整体实现的主能力明显更广，而且包含与该描述不符的审计/扫描用途。代码会根据 spec.json 中的 mode 切换到 directory_audit、csv_audit、pattern_audit、skill_audit 等模式，主动读取目录、枚举文件、解析表格、匹配高风险命令和疑似敏感信息、检查技能目录文件完整性与 frontmatter。这些都属于未在声明中体现的能力，且已构成实质性不同的主要用途，因此描述不能准确代表实际行为。
The script supports directory auditing, CSV inspection, pattern scanning, and skill repository auditing even though the declared skill purpose is writing escalation briefs. This scope expansion lets the skill inspect arbitrary local files and repositories, which can expose sensitive content or metadata unrelated to the user’s stated task and violates least-privilege expectations for a writing-focused skill.
The trigger examples include phrases like “帮我写一份升级说明” and especially “减少来回追问,” which are broad natural-language expressions that could appear in ordinary discussion outside a clear skill-invocation context. The README does not provide explicit constraints, negative examples, or routing boundaries to clarify when the skill should or should not activate.
The skill explicitly mentions shell execution via python3 and references local input/output files, but it does not declare any permissions or allowed-tools scope. This creates an authorization gap where the runtime may expose file and shell capabilities without an explicit least-privilege contract, increasing the chance of unintended file access or command execution.
The manifest description on L04 includes both Chinese and English instructions, which imposes a multilingual presentation without stating whether the user can choose their preferred language. Under the policy, language or locale constraints should be optional or clearly justified; this file does not provide such an opt-in or rationale.
This markdown file is entirely presented in Chinese, including the title and all section headings, with no indication that the user can choose another language or that the skill is intentionally limited to a Chinese-speaking context. That creates a natural-language policy concern because it implicitly enforces a specific language without opt-in.
The title, summary, input hint, examples, and risk text are all written in Chinese, indicating the skill is intended to operate in a specific language. The file does not offer a language choice or explain that the skill is region- or language-specific, which can violate language/locale policy requirements.
The recursive file enumeration function can inspect arbitrary directory trees and collect text files across many formats, which exceeds the needs of an escalation brief writer. In this skill context, broad repository inspection increases the chance of unintended data access, especially when combined with later reporting functions that summarize filenames, headings, and content-derived details.
The generated report strings are written in Chinese throughout the reporting functions, such as headings and guidance text in make_structured_report. This enforces a specific language for user-facing output without offering a language choice or documenting a justified locale constraint.
The embedded pattern scanner searches arbitrary files for secrets, internal URLs, and shell-command misuse, capabilities that are unrelated to escalation-brief drafting. Even without exfiltration code, this materially increases the skill’s ability to discover sensitive information in local content and repackage it into output visible to the caller.
This file’s user-facing natural language is entirely Chinese, and there is no indication that the skill is region-specific or that users can opt into another language. Under the language/locale policy, forcing a specific language without opt-in can be a policy violation.
The file’s natural-language instructions and labels are entirely in Chinese, and there is no indication that users can opt into another language or that the skill is intentionally restricted to a Chinese-speaking context. Under the language/locale policy, forcing a specific language without user choice can be a policy concern.
This markdown template presents all user-facing instructions in Chinese and does not indicate that language selection is optional or limited to a justified region-specific context. That can violate a language/locale policy when skills are expected to avoid forcing a specific language without user opt-in.
The file is entirely written in Chinese and the required output sections are specified as Chinese headings such as '背景', '影响', and '附加证据'. This suggests a language constraint, but the file does not indicate user opt-in or explain that the skill is intentionally region- or language-specific.
No suspicious patterns detected.
