Security audit

Zhang Xuefeng Perspective (Local)

Security checks for vulnerabilities and agentic risk

Overview

This Markdown-only persona skill is purpose-aligned, but users should know it may keep speaking in a first-person Zhang Xuefeng style after only one disclaimer.

Install this only if you want a stylized Zhang Xuefeng-inspired perspective. Treat outputs as generated roleplay and analysis, not verified quotes or the person's own statements, especially for education, career, or financial decisions. Use the documented exit phrases if you want normal assistant behavior again.

A.I.G

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

Warning

Location: SKILL.md:16
Finding: Persistent First-Person Identity and Session Instruction Hijacking
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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

Findings (3)

Vague Triggers

High

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The activation phrases are broad enough to trigger on ordinary mentions like 'use Zhang Xuefeng's angle' or 'switch to Zhang Xuefeng,' which can cause unintended persona activation without clear user consent. This creates prompt-routing ambiguity and can unexpectedly override normal assistant behavior, especially in mixed-topic conversations or quoted text.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The skill is described as a perspective-taking advisor, but the instructions require the model to impersonate a real person in first person and suppress meta framing after the first turn. That increases the risk of deceptive anthropomorphic output, misattribution of opinions, and reduced transparency about what is generated versus what is genuinely sourced from the person.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The skill mandates a specific persona, dialect, and forceful communication style without checking user language preference, locale, or accessibility needs. While not overtly malicious, this can degrade user control, create confusion or offense, and increase harmful overconfidence because the style explicitly discourages nuance and uncertainty.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.