Security audit
Security checks for vulnerabilities and agentic risk
This skill is a local Chinese-language review/feedback summarizer that reads user-supplied files and produces reports, with no evidence of network access, persistence, or destructive behavior.
Install only if a Chinese-language local review/feedback reporting workflow fits your use case. Provide only files you intend the skill to read, redact personal customer information first, and review generated output before using it in marketing or operational decisions.
描述聚焦于“从用户评论和反馈中提炼营销洞察”，但代码没有任何专门针对评论、评价、反馈、情感、卖点/痛点/反对意见提取的逻辑。相反，它是一个可配置的通用报告器/审计器：可扫描目录中的文本文件、统计扩展名、读取 CSV 字段、检测 shell/secret/private URL 等风险模式、检查 Skill 目录必备文件和 frontmatter。虽然在某些 mode 下，structured_brief 会把输入文本按 section 生成模板化报告，表面上可能被用于评论整理，但这只是泛化的分段输出，不具备所宣称的特定分析能力。其主要用途与声明明显不一致，且包含多个未声明的审计/扫描能力，因此应判定为描述与行为不匹配。
The script supports multiple generic audit modes such as directory, CSV, pattern, and skill auditing that materially exceed the declared review-mining purpose. This capability expansion can enable unintended inspection of arbitrary local files and repositories, increasing the chance of unauthorized data exposure or misuse in environments where the skill is assumed to process only customer feedback.
The skill advertises executable capabilities via python3 and references local file inputs/outputs, but it does not declare any explicit tool scope such as allowed tools or permissions. That creates an authorization gap: an agent or reviewer cannot easily constrain the skill to the minimum necessary capabilities, increasing the chance of unintended file access or shell execution if the skill is invoked in a permissive runtime.
The entire template, including the title and all section headings, is written in Chinese with no indication that users may choose another language. Under the policy criteria, a skill that imposes a specific language without user opt-in can be a natural-language policy violation.
The generated report strings in make_structured_report are hard-coded in Chinese, including titles, summaries, and guidance text. This enforces a specific language for users regardless of their locale or preference, and the file provides no option to choose another language.
Multiple reporting functions, including directory_report, csv_report, pattern_report, parse_frontmatter error messages, and skill_audit, emit Chinese-only natural-language output. Because all major output modes are fixed to one language and no opt-in or locale selection exists, this violates the language-choice policy.
The pattern scanner searches arbitrary files for secrets, private URLs, and dangerous shell snippets, which is outside the advertised review-mining workflow. In a broader agent environment, this can be repurposed to enumerate sensitive material from local content and surface it in generated reports, even if values are only partially masked.
The skill can audit repository structure and parse SKILL.md/frontmatter for other skills, which is unrelated to extracting insights from reviews. This broadens the skill into a general local reconnaissance utility, creating unnecessary access to project metadata and file inventories that may reveal internal structure or sensitive operational details.
The manifest description and core instructions are written partly in Chinese and partly in English, but the file does not state that the user may choose their preferred language. This can violate a language/locale policy when a skill imposes or assumes a language without explicit opt-in.
The entire skill example is written in Chinese and includes output instructions such as '使用清晰标题', but it does not state that language selection is optional or user-driven. Under the policy, a language constraint should either offer user choice or be clearly justified as region-specific.
The markdown example is entirely written in Chinese and presents a fixed output structure in that language, with no indication that users can choose another language or locale. Under the policy rule, forcing a specific language without user opt-in is a natural-language policy concern.
This markdown file presents all user-facing test instructions in Chinese, which can amount to forcing a specific language without user opt-in. The file does not state that the skill is region-specific or offer any language/locale alternative, so it may violate language or locale policy requirements.
No suspicious patterns detected.
