Back to skill

Security audit

段永平视角

Security checks for vulnerabilities and agentic risk

Overview

This is a persona-style investment framework skill with no executable code, install actions, persistence, credential handling, or hidden data access.

Install this only if you want responses framed through a Duan Yongping-style value-investing lens. Treat the output as perspective and analysis, not financial advice, and be aware that broad trigger phrases may steer ordinary investment questions into 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 (2)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill is written to enforce a Chinese-language and persona-specific response style without presenting language choice or user consent. This can reduce user control, cause inaccessible or confusing output for non-Chinese users, and create mismatches between user intent and assistant behavior, especially when the skill is activated indirectly.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger phrases are broad and include common terms like '价值投资' and '长期持有', which can cause the skill to activate in conversations where the user did not explicitly request this persona. Unintended invocation can override normal assistant behavior, steer answers into a fixed investment framework, and increase the risk of unsuitable or misleading financial guidance in unrelated contexts.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.