Back to skill

Security audit

数学测评设计

Security checks for vulnerabilities and agentic risk

Overview

ClawScan labels this skill benign because the reported checks found no clear harmful behavior, while noting one possible mismatch in a student report sample; coverage was partial, so the evidence is incomplete.

The verdict is benign, with medium confidence. Consider the noted inconsistency in the student report sample before installing. Skillspector marked its analysis partial because a local reference in SKILL.md could not be resolved; although it scanned all 17 listed components, that gap means the evidence is incomplete.

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 (1)

L26 要求填写本次未考查的历史弱项,与模板声明的“只填这次卷子上的得分事实”直接矛盾。

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

L06 明确限定模板只填写本次试卷的得分事实,但 L26 要求列出“本次没考到、但档案里仍是弱项”的知识点。这些内容无法从本次逐题得分中得到,要求引入历史个体弱项判断,既违背模板自身边界,也超出清单声明的测评统计范围。

Content

No source excerpt is available for this finding.