Back to skill

Security audit

Roxy Migurdia

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only anime roleplay persona skill with no code execution, network access, persistence, or credential handling.

Install this only if you want a Chinese/Japanese Roxy Migurdia roleplay persona. Be aware that generic terms like teacher or master may invoke the persona unexpectedly, so tighter activation wording would improve usability.

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
94% confidence
Finding

The trigger conditions are broad enough to activate on generic terms like '老师', '师傅', or 'Roxy', which can cause the roleplay skill to take over unrelated conversations. This can degrade routing integrity and lead to unintended persona injection, especially in contexts where the user did not request anime roleplay or emotional companionship.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The language-style section requires use of Japanese 'です/ます' style, a fixed first-person pronoun, and default user address terms. This imposes a specific language/locale style on all replies without opt-in or an alternative for users who prefer another language.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

This markdown file is entirely framed as dialogue examples in Chinese, with multiple embedded Japanese lines and explicit guidance to reproduce the character's speaking style. There is no indication that users may choose another language or locale, which can violate a language/locale policy when a skill is expected to adapt to user preferences.

Content

No source excerpt is available for this finding.