Security audit
Security checks for vulnerabilities and agentic risk
This is a local metrics-catalog helper that is mostly purpose-aligned, with some dormant generic audit code users should be aware of.
Install only if you want a Chinese-first local helper for organizing metric definitions. Keep inputs scoped to the files you intend to summarize, and be aware that the bundled script contains unused audit modes that could scan broader local content if someone modifies the spec configuration.
声明描述的是一个面向指标定义整理的专用能力，但代码并未体现任何与‘指标口径、公式、归属、例外情况、常见误用’有关的领域逻辑。相反，代码的核心行为由 resources/spec.json 中的 mode 决定，可执行多种通用审计/扫描任务：扫描目录文本文件、统计扩展名、抽取 Markdown 标题、读取 CSV 生成字段概览、对文件内容执行风险模式匹配、检查 Skill 目录文件完整性和 SKILL.md frontmatter。虽然 structured_brief 模式会把输入整理成分节报告，表面上可用于某些文档梳理，但它只是通用模板化摘要器，不是专门的指标目录整理工具。因此代码的主要用途与声明存在实质偏差，并包含多项未声明的审计/扫描能力。
The script implements broad directory, CSV, pattern-scanning, and skill-audit behaviors that materially exceed the stated purpose of cataloging metric definitions. This scope mismatch is dangerous because it enables filesystem inspection and security-style analysis over arbitrary user-supplied paths, increasing unintended data exposure and repurposing risk in a skill that users would not expect to have audit capabilities.
The built-in risky-pattern scanner searches arbitrary files for secrets, private URLs, and shell-danger patterns, which is unrelated to a metric catalog skill. In context, this creates an unnecessary content-inspection capability that can surface sensitive snippets from local repositories and turns the skill into a lightweight security scanner without clear user expectation or need.
The trigger examples "整理这批指标定义" and especially "统一口径和计算方式" are natural phrases that could appear in general collaboration without clearly signaling this specific skill. The README does not provide constraints, explicit invocation patterns, or negative examples to distinguish when the skill should or should not activate.
The skill advertises executable capability via python3 and references reading templates/specs and writing output, but it does not declare any explicit tool scope such as allowed tools or permissions. That creates a least-privilege gap: an agent or reviewer cannot reliably tell what filesystem or shell access is intended, increasing the chance of overbroad execution in a supposedly documentation-oriented skill.
The manifest description forces mixed-language instructions (Chinese and English) as the default presentation, but does not state that the user can choose their preferred language or locale. This can violate a language/locale policy when skills are expected to respect user language preference unless explicitly justified.
This markdown file presents all user-facing headings and instructions exclusively in Chinese, and it does not indicate that the skill is region-specific or that users may choose another language. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.
Multiple generated user-facing report strings are hard-coded in Chinese, such as headings and guidance text, with no option for the user to select language or locale. This creates a natural-language policy issue because the skill enforces a specific language without opt-in or justification.
The repository/skill packaging audit capability checks for project files and parses SKILL.md frontmatter, which is outside the declared analytics catalog use case. While less severe than the pattern scanner, it still broadens the skill into filesystem and repository inspection, creating unnecessary access to local project structure and metadata.
This markdown file’s user-facing instructions and expected outputs are fully Chinese-language, and there is no indication that users can opt into another language or that the skill is intentionally limited to a Chinese-speaking context. Under the policy, forcing a specific language without user choice is a natural-language policy violation.
The file presents the skill name, usage, triggers, and warnings only in Chinese, which effectively forces a language/locale choice in its natural-language interface. There is no indication that users may choose another language or that the skill is intentionally limited to a Chinese-speaking context.
The file is entirely written in Chinese and does not indicate that language choice is optional or limited to a justified region-specific context. Under the policy for natural-language violations, a skill that effectively constrains interaction to a specific language without opt-in can be considered a locale policy issue.
This markdown template forces a specific language for all headings and instructions, which can violate language/locale policy when no user opt-in or justification is provided. The file contains no indication that the skill is region-specific or that users may choose another language.
No suspicious patterns detected.
