Back to skill

Security audit

历史材料解析题教练

Security checks for vulnerabilities and agentic risk

Overview

This is a self-contained Chinese K12 history tutoring skill with disclosed, opt-in profile and reminder handoffs; it should be localized before use outside mainland China.

Use this skill for Simplified Chinese, China-curriculum history tutoring. Before deploying in the United States or another region, localize curriculum assumptions, privacy/guardian-consent rules, and crisis referral resources; keep profile sharing and reminders enabled only with explicit student or guardian consent.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (7)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The entire reference guide is written as a Chinese-only instructional resource for the skill, with no indication that users may choose another language or locale. Under the policy rule, forcing a specific language without user opt-in is a natural-language policy concern unless the constraint is explicitly documented and justified as region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The entire protocol is written only in Chinese and includes locale-specific operational guidance centered on China Mainland resources, without stating that users can choose another language or locale. This can violate language/locale policy when the shared skill is used in broader deployments, especially since it is described as applying repository-wide.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The schema's human-readable note and description are written entirely in Chinese and the enumerated values throughout the file are Chinese-only, which effectively fixes the skill's language/locale context in natural-language metadata. For SQP-3, this is a policy concern because the file does not offer a language choice, opt-in, or justification that this schema is intentionally region-specific.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The schema explicitly distributes and permits the wrong_answer_handover path even though the skill metadata says this skill does not handle initial wrong-answer intake or 28-day counting. That mismatch can let the skill emit or accept authority-bearing wrong-answer records outside its intended scope, creating unauthorized cross-skill data flow and integrity problems in learner records.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The schema grants reminder_enqueue capability to this skill, which expands behavior from analysis/tutoring into user-affecting outbound actions. Even with a consent field present, exposing an unnecessary capability increases the chance of overreach, consent misuse, or prompt-driven triggering of reminders unrelated to the skill's stated purpose.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The schema title/description and related natural-language fields are written to require Chinese-language protocol terms such as Chinese enum labels and descriptions, with no indication that users or integrators may choose another language. This is a natural-language locale constraint embedded in a config/schema file, and the file does not document a justified region-specific limitation or offer opt-in choice.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

L11 明确写明“本库按中国大陆的中文 K12 场景设计”,并指定示例话术为简体中文、求助渠道默认中国大陆。这属于自然语言层面的语言/地域约束;虽然文件后续说明了跨地区部署需先做本地化处理,但默认约束本身仍构成需要关注的 locale policy 表述。

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.