Security audit
Security checks for vulnerabilities and agentic risk
This is a local Chinese-language webinar repurposing skill that reads user-provided material and can write a chosen output file; no active exfiltration, persistence, or unsafe execution was found.
Install this only if you want a Chinese-language webinar content repurposing helper. Provide only material you are comfortable processing locally, avoid feeding secrets or confidential transcripts without review, and expect draft Markdown output rather than any automatic posting or sending. The publisher should remove or separate the dormant audit/scanning code to make the package easier to trust.
声明描述的是一个面向 webinar 内容再利用的内容生成/拆解工具。但代码本身并未实现任何 webinar 专属处理逻辑，也没有针对“回放介绍、短文案、社媒提纲、FAQ、后续邮件”的生成流程。相反，它从 resources/spec.json 读取配置后，按 mode 执行多种通用审计/报告任务：扫描目录中的文本文件、汇总扩展名、读取 CSV 字段、用正则匹配潜在危险命令或疑似密钥、检查 Skill 工程文件完整性与 frontmatter。这些能力与声明的主要用途明显不同，且包含未声明的代码/目录审计与模式扫描能力。因此描述与实际行为存在实质性不匹配。
The code can recursively inspect arbitrary directories, list files, sample markdown headings, and summarize local content structures, which exceeds what a webinar repurposing tool should need. In a mislabeled skill, this broad filesystem visibility can be abused to inventory and extract contextual information from unrelated user data.
The script’s primary behaviors are directory auditing, pattern scanning, CSV inspection, and skill-package validation, which are materially unrelated to the declared webinar-repurposing purpose. This creates a capability mismatch that enables broad local file enumeration and content inspection under a benign-seeming skill label, increasing the risk of unintended data discovery or misuse.
The embedded regex engine scans arbitrary file content for secrets, private URLs, and shell-danger patterns, which is not justified by a webinar content workflow. In this context, the feature can expose sensitive material from local files and gives the skill surveillance-oriented behavior inconsistent with user expectations.
L33-L35 仅给出“把 webinar 内容拆成多平台素材”“生成 FAQ 和邮件草稿”这类能力示例，但没有说明是否为唯一触发短语、适用前提，或哪些相似请求不应触发该技能。对 markdown 技能说明而言，这种描述更像宽泛能力声明，缺少明确约束，容易与常见内容创作请求重叠。
The skill advertises shell-enabled execution 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 an authorization ambiguity where a host agent may grant broader file or shell access than users expect, increasing the chance of unintended file access or command execution.
The manifest description and core skill identity are written as instructions in Chinese, presenting the skill as operating in that language by default. There is no stated user opt-in, alternative locale handling, or language choice, which can violate language/locale policy requirements for skills intended for general use.
This markdown template is entirely written in Chinese and presents the output structure only in that language, with no indication that users may choose another language or that the skill is intentionally limited to a Chinese-language context. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation.
Multiple generated user-facing report strings are hard-coded in Chinese, which imposes a specific language/locale on all users. The file provides no opt-in, configuration, or justification that this skill is intended only for a Chinese-language environment.
The smoke test file is entirely written in Chinese and requires generated output sections such as '核心观点' and '回放介绍', which implies the skill expects or enforces a specific language/locale. There is no indication that users can choose another language or that the locale restriction is justified as region-specific.
The file content is entirely in Chinese and includes instruction-like text, which implies the skill may operate in a fixed language context without stating that the user can choose another language. Under the policy rule, forced language or locale behavior should be explicitly optional or justified.
This markdown file presents all headings and instructions in Chinese, which can amount to a language/locale policy issue if the skill forces a specific language without user opt-in. The file does not indicate that Chinese is optional, user-selected, or required for a justified region-specific use case.
No suspicious patterns detected.
