Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent teacher-facing English listening lesson design skill with disclosed classroom data use and no executable or hidden install behavior.
Install this for Chinese K12 teacher workflows where class assessment and listening records may be available to the platform. Operators outside Mainland China should localize curriculum assumptions, legal guidance, and crisis resources before student-facing use, and teachers should keep the confirmation and consent controls enabled before writing items or student-profile updates.
The title and all instructional content are written entirely in Chinese, and the file gives no indication that users may choose another language or that Chinese is a required locale for a region-specific deployment. Under the policy, forcing a specific language without opt-in is a natural-language policy concern.
This markdown file presents all instructions and guidance exclusively in Chinese, and it does not indicate that the user opted into that language or that the skill is intentionally limited to a Chinese-speaking context. Under the policy, forcing a specific language without user choice can be a natural-language policy violation.
SQP-3 applies to all file types and covers natural-language policy violations such as forcing a specific language without user opt-in. This sample presents all instructional content only in Chinese and does not indicate any user language selection or justified locale restriction.
This markdown file presents all instructions and examples exclusively in Chinese, with no indication that the user can choose another language or that the locale restriction is intentional. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.
The schema copy exposes multiple assessment-analysis datasets (itemScores, itemStats, weaknessRank, reviewPlans) to a skill described primarily as listening lesson design. Even where some are read-only, over-broad data exposure increases attack surface and enables unnecessary access to student performance analytics, making data misuse, prompt leakage, or unintended cross-function behavior more likely.
The scoped schema explicitly grants this listening-lesson design skill write access to examBlueprints, allowing it to create or modify assessment structures outside its described lesson-planning workflow. In an agentic system, unnecessary write capability expands the blast radius: a compromised or misaligned skill could alter exams, inject unverified items, or interfere with downstream assessment processes.
The file is entirely written in Chinese and includes normative instructions for SKILL behavior, such as user-facing applicability and reminder rules, but does not state that language/locale is optional or limited to a China-specific deployment. Under the policy, forcing a specific language or locale without user opt-in is a natural-language policy violation.
The document is entirely written as a normative shared policy file for skills, and several directives and examples assume Chinese-language interaction by default, including required one-sentence wording and skill behavior. Because no user language choice, opt-in, or documented locale limitation is provided, this can violate a language/locale policy requiring flexibility or explicit justification.
The document states that the repository is designed for '中国大陆的中文 K12 场景' and defaults safety-help channels, curriculum alignment, and legal assumptions to Mainland China. This is a natural-language locale policy constraint, and the file does not offer user opt-in or selection within the skill itself.
The file is entirely written as normative instructions in Chinese and presents itself as the sole source of terminology and enums for the repository. There is no indication that users or downstream skill authors may choose another language or locale, which constitutes a language policy constraint without opt-in.
No suspicious patterns detected.
