Security audit
Security checks for vulnerabilities and agentic risk
The skill is coherent and not malicious, but it can automatically change project code over a broad default scope without requiring clear confirmation.
Install only if you are comfortable with an agent editing accessibility-related code. Use explicit target paths, ask for report-only or dry-run behavior first, and review diffs before accepting changes, especially in large repositories.
The README states that the skill will automatically fix accessibility issues but does not clearly warn that project files may be changed. In a tool that can perform edits, lack of upfront disclosure can lead to unexpected modifications, especially if the skill is triggered broadly or used in a large codebase.
The trigger phrases are broad, generic accessibility terms that overlap with ordinary user requests, so the skill may activate when the user did not explicitly intend to invoke it. Because the skill can scan and automatically modify files, unintended invocation increases the chance of surprise code changes or noisy edits in unrelated contexts.
整个技能说明以中文固定呈现，且未声明可根据用户偏好切换语言或接受其他语言输入/输出。按规则，强制特定语言但未提供选择或明确、合理的地区性约束，属于自然语言策略违规。
L011-L012 使用“无障碍改进”“兼容性问题”等宽泛表述作为触发条件，这些短语在日常开发讨论中覆盖面很广，且未限定必须与 React/TSX 审查或自动修复相关。文档也没有提供排除条件或负例，导致技能何时应激活、何时不应激活不够明确。
The skill explicitly states it will scan the current directory or specified files and automatically apply fixes, but it does not require an explicit confirmation or warn about the risk of bulk modification. In an agentic environment, this can lead to unintended mass edits, corruption of unrelated files, or unsafe changes being applied beyond the user's intended scope.
The usage text prominently states that the skill will automatically scan and repair code, and later run lint checks, but does not provide an upfront warning in the behavior description that files may be modified or commands may be executed. In an agent setting, this can create unsafe expectations and cause users to invoke the skill without realizing it may perform write actions or environment-dependent verification steps.
The skill can be triggered by very broad natural-language phrases such as general accessibility review requests, which increases the chance of accidental invocation in normal conversations. Because this skill is described as automatically scanning, modifying files, and later running checks, a false trigger could lead to unintended code changes or workflow disruption without sufficiently explicit user intent.
The tagline and description are entirely in Chinese, and the file provides no indication that language is selectable or that the skill is intended only for a Chinese-speaking or region-specific audience. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation.
This markdown file contains natural-language content exclusively in Chinese, and there is no indication that users may choose another language or that the file is intentionally region-specific. Under the policy rule for language/locale constraints, forcing a single language without opt-in can be a policy violation.
Nearly all user-facing instructions and invocation guidance in the README are presented only in Chinese. Because the file does not offer an alternative language or indicate that Chinese is an intentional locale-specific constraint, it may violate language-choice policy expectations.
文件标题、正文、命令示例及生成内容示例均固定为中文，没有说明用户可选择其他语言，也未将该限制解释为特定地区或组织要求。这可能构成语言/locale 约束未获用户选择的自然语言策略问题。
This file’s instructional content, examples, and report template are all presented only in Chinese, with no indication that the user can opt into another language or that the skill is intentionally region-specific. Under the policy, forcing a specific language without user choice or documented justification is a natural-language policy concern.
This markdown file presents all instructions, examples, and guidance exclusively in Chinese. Under the policy rule for natural-language violations, forcing a specific language without user opt-in or a documented locale-specific justification is a reportable issue.
No suspicious patterns detected.
