Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed job-matching skill that reads user-provided resume and job data and writes matching results, with no evidence of hidden code, exfiltration, or persistence abuse.
Before installing, be comfortable with the agent reading the resume/JD files or Feishu table links you provide and writing priority fields or separate result files. Clarify where any reused candidate profile or matching configuration will be stored, and review Feishu updates before using it on shared business tables.
The primary descriptive text is presented as Chinese-only and identifies the skill as dedicated to a specific environment, but there is no statement that users may choose another language or that Chinese is a required locale for compliance or regional operation. Under the policy, forcing a specific language without opt-in is a natural-language policy concern.
The README states that the agent will automatically trigger this skill for broad recruiting-related requests such as resume matching, JD matching, and job filtering. Overly broad activation criteria can cause unintended invocation on common user requests, leading the agent to process resumes, preferences, and job data without sufficiently explicit user intent or a narrower routing condition. In this context, the skill handles sensitive employment-related data, which makes accidental activation more consequential.
The skill title and all operational instructions are presented in Chinese, and the document does not state that language selection is optional or region-specific. Under the language/locale policy, forcing a single language without user opt-in is a natural-language policy concern.
The documentation explicitly states '全链路 Agent 推理，零脚本' at L016 and repeats '不使用脚本' at L311, suggesting a purely reasoning-based flow without operational side effects. However, the same file defines writing results to Feishu bitables via batch_update and creating new local CSV/Excel output files, which contradicts the no-script/no-automation claim at the intent level.
The trigger condition is very broad ('when the user asks for resume matching, JD matching, or job screening'), which can cause the skill to activate on loosely related recruiting conversations without clear user intent. In a skill that processes resumes, local files, spreadsheets, and remote tables, over-triggering can lead to unnecessary access to sensitive employment data or unintended write operations.
The document's instructions and labels are fully in Chinese, which effectively forces a specific language for users of the skill. Under the policy, locale or language constraints should either be optional for the user or clearly justified as region-specific.
This markdown file presents all operational rules and user-facing interaction examples entirely in Chinese, including the prescribed agent prompt at L074-L079. The file does not indicate that language is optional, configurable, or limited to a justified region-specific context, which can violate a language/locale policy requiring user choice.
This markdown file is written primarily in Chinese and includes English scoring/status terms such as "medium" and "low," but it does not indicate that the skill is intentionally Chinese-only or offer users a language choice. Under the policy rule for language/locale, forcing a specific language without opt-in can be a natural-language policy concern.
This markdown file uses Chinese throughout for headings, field names, instructions, and examples, but does not indicate that the skill is Chinese-only or provide an opt-in language choice. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale constraint is explicitly documented and justified.
No suspicious patterns detected.
