Back to skill

Security audit

化学用语过关训练设计

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Chinese K12 chemistry-teaching skill with scoped classroom data use and no executable or hidden install behavior.

Install this for Chinese-language chemistry classroom workflows where teachers review outputs before use. Do not deploy it directly to students outside mainland China unless crisis contacts, curriculum alignment, legal defaults, and language expectations are localized first.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The schema note is written to mandate Chinese-language field usage and references Chinese-only documentation and terminology, which indicates a fixed language/locale expectation. Under the policy, forcing a specific language without offering choice or documenting a justified region-specific scope is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

This file contains user-facing operational instructions only in Chinese, with no statement that the language is region-specific or that another language can be selected. Under the language/locale policy, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This file contains user-facing instructional policy text exclusively in Chinese, but it does not indicate that Chinese is optional or that the skill is limited to a Chinese-language context. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file hard-codes a China-mainland K12 locale as the default operating context and warns against direct student use elsewhere unless operators localize emergency contacts, curriculum alignment, and legal defaults. This is not overtly malicious, but it can still create a safety and compliance vulnerability if the skill is deployed in other regions without explicit user locale detection or opt-in, leading to wrong crisis resources, age-of-consent assumptions, or educational guidance.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The description lists trigger examples such as “过关卡怎么设计”, which is broad and not tightly anchored to chemistry notation drills. A generic request about designing a pass card could overlap with many unrelated teaching-planning situations and may cause unintended invocation despite the later scope exclusions.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.