Back to skill

Security audit

胡田 OPC导师 范仲淹 心智模型

Security checks for vulnerabilities and agentic risk

Overview

This is a Chinese-language persona and advice skill based on Fan Zhongyan, with no code execution, credential access, persistence, or hidden system changes.

Install this if you want Chinese-language, value-laden coaching inspired by Fan Zhongyan. Review the broad activation examples and limit use to explicit requests if you do not want everyday career, team, or life conversations framed through this persona.

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 (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill recommends invocation based on very broad personal and professional situations like setbacks, low points, or team difficulties, which could cause the agent to apply this persona in many ordinary conversations without clear user opt-in. That creates scope creep and increases the chance of unsolicited behavioral steering or inappropriate responses framed through a specific ideological lens.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The example trigger phrases are open-ended and easy to match semantically, encouraging activation from loosely related user inputs rather than deliberate consent. In a persona-guidance skill, this can lead to over-activation and unrequested value-laden advice, especially in sensitive life or work decisions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The entire skill content is presented exclusively in Chinese, with no indication that language choice is optional or that the skill is limited to a Chinese-language or region-specific use case. Under the policy, natural-language files that impose a specific language without user opt-in should be flagged.

Content

No source excerpt is available for this finding.