Security audit
Security checks for vulnerabilities and agentic risk
This skill is a local checklist generator with no evidence of hidden network access, persistence, credential use, or unsafe automatic actions.
Install only if you are comfortable with a Chinese-language local helper that processes files you point it at and can write a report to a chosen output path. Treat the bundled audit functions as unnecessary dormant code; avoid editing spec.json to enable them unless you intentionally want local scanning behavior.
声明描述的是一个面向 PRD/接口文档/需求规格的清单拆解工具，核心用途应是生成验收、联调、测试、上线等工作清单。实际代码虽然在某些 mode 下可以做“structured_brief”式结构化摘要，勉强接近清单整理，但其主要实现是一个通用本地报告/审计脚本框架：它会读取本地文件或目录，枚举文本文件，统计扩展名，抽取 Markdown 标题，分析 CSV 字段，执行正则模式扫描（如 curl|bash、rm -rf、secret、private URL），以及检查 Skill 工程文件完整性与 SKILL.md frontmatter。这些能力明显超出并偏离了“需求拆解为验收/测试/上线清单”的已声明范围。代码还访问并分析本地目录与多类文件资源，而声明未体现这类审计/扫描能力。因此应判定为描述与实际行为存在实质性不匹配。
The script implements multiple behaviors unrelated to the declared spec-to-checklist purpose, including directory auditing, pattern-based security scanning, CSV auditing, and skill package auditing. This expands the skill’s effective capability and trust boundary, making it able to inspect local repositories and surface sensitive content in contexts where users would expect only checklist generation.
The regex scanning capability searches arbitrary files and directories for secrets, internal URLs, and risky shell patterns, which is not justified by the skill’s stated purpose. In a local agent environment, this creates a data-discovery mechanism that can expose sensitive snippets from unrelated files and encourage users to run a broader scanner than they intended.
The README presents all user-facing instructions, examples, and warnings only in Chinese. Under the stated policy, forcing a specific language without user opt-in or documented justification is a natural-language policy violation.
The skill advertises local resources and explicitly permits shell execution via python3 .../scripts/run.py, but it does not declare any explicit tool scope such as allowed-tools or permissions. That mismatch weakens least-privilege controls and can let the agent invoke file read/write or shell capabilities without clear user-visible boundaries, increasing the chance of unintended file access or command execution.
The description is written partly in Chinese and partly in English, which effectively imposes multilingual output/interaction expectations without stating that users may choose their preferred language. Under the policy rule, forcing a specific language or locale without opt-in is a natural-language policy concern.
Multiple user-facing report strings are emitted only in Chinese, such as headings, summaries, and next-step instructions. This imposes a specific language/locale on all users with no opt-in or documented justification, which matches the language/locale policy violation category.
The skill_audit functionality inspects skill package structure and parses SKILL.md frontmatter, which is outside the declared checklist-generation behavior. While less severe than secret scanning, it still introduces undeclared repo inspection capability and can be used to enumerate package internals or analyze neighboring content without clear user expectation.
This markdown file presents all user-facing audit content in Chinese and does not indicate that users may choose another language. The policy explicitly calls for flagging language or locale constraints when a skill forces a specific language without user opt-in.
This markdown example output forces a specific language/locale presentation throughout the file, and there is no natural-language indication that users can choose another language. Under the policy rule, fixed language output without user opt-in can be a locale-policy violation.
The manifest presents the skill name, category label, and summary entirely in Chinese, which can constitute a language/locale policy issue if users are not given an explicit choice or opt-in. There is no indication in this file that the skill is region-specific or that an alternative language is available.
The entire template, including its title and instructions, is written only in Chinese and provides no indication that language choice is optional or limited to a justified region-specific context. This can violate language/locale policy guidance because it implicitly constrains skill output language without user opt-in.
The entire skill test document is written only in Chinese, which can constitute a language policy violation when no user choice or opt-in is offered. The file does not indicate that the skill is region-specific or otherwise justified to require Chinese-only operation.
No suspicious patterns detected.
