Tp2
High
- Category
- MCP Tool Poisoning
- Confidence
- 85% confidence
- Finding
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
- Content
Security audit
Adcote Kede-浦江科德融合部简介
Security checks for vulnerabilities and agentic risk
Overview
This is a self-contained school information writing skill with no executable code, persistence, or data access, though some activation phrases are broader than ideal.
Before installing, consider narrowing the trigger phrases so the skill activates only when the user mentions 科德高中, 浦江县科德高级中学融合部, or Adcote School Zhezhong Campus. Users should also treat policy, fee, admissions, and university application details as needing confirmation from official current sources, as the skill itself advises.
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
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 89% confidence
- Finding
The trigger list includes broad terms such as '浦江', '国际部', '学校介绍', and '课程说明', which can match many unrelated school or regional conversations. This can cause the skill to activate outside its intended scope, leading to incorrect routing, irrelevant responses, or accidental overshadowing of more appropriate skills in user interactions.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 89% confidence
- Finding
The skill explicitly targets '40—55 岁浙江家长' and prescribes a localized communication style without indicating that the assistant should first confirm the user's preferred language, region, or audience. This can cause the agent to override user intent, reduce inclusivity, and produce inappropriate or biased outputs for users outside that demographic.
- Content
Static analysis
No suspicious patterns detected.