Back to skill

Security audit

Yu Liang Perspective V2

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed conversational persona skill with no code, tool access, credential handling, network use, or system persistence, though its roleplay framing could confuse users if triggered unintentionally.

Install this only if you want a Chinese-language business strategy persona that frames answers around Yu Liang-style themes such as survival, cash flow, prudence, and long-termism. Be aware that generic phrases like “活下去” or “现金流为王” may trigger the persona, and outputs should not be treated as real quotations or advice from Yu Liang unless independently sourced.

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
Findings (1)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:19
Finding

Persistent Named-Person Persona and Response Instruction Hijacking

Content
View full analysis
Remediation
View remediation
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)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger list includes broad phrases such as “活下去” and “现金流为王” that can plausibly appear in ordinary business discussions, causing the skill to activate when the user did not intend to invoke this persona. Unintended activation can override the assistant’s normal behavior and steer responses into roleplay or constrained advice, reducing reliability and user control.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The skill instructs the agent to respond directly as 郁亮 after activation and prescribes first-person identity and stylistic constraints without asking for fresh user consent at response time. This is risky because persona enforcement can obscure that the model is roleplaying and may cause users to receive framed or authoritative-sounding answers they did not explicitly choose in the current turn.

Content

No source excerpt is available for this finding.