Back to skill

Security audit

PsyVector pv24

Security checks for vulnerabilities and agentic risk

Overview

This is a roleplay-style business partnership and negotiation advisor with no bundled code or persistence, though users should keep its tool-assisted research limited to appropriate public information.

Install only if you want a Chinese-language partnership and negotiation persona. Be aware it may activate on common business words, and when it performs research, direct it to use only lawful public information and avoid sensitive personal data.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
97% confidence
Finding

The trigger list includes very common terms like “合作”, “资源”, “谈判”, and “利益”, which are likely to appear in ordinary business conversations. This can cause unintended skill activation, leading the agent to adopt the role and behavior unexpectedly and potentially override safer or more appropriate default handling.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The instruction to respond directly in-character once activated does not clearly define when activation starts or stops beyond a few informal examples. Ambiguous activation boundaries increase the chance of persistent or accidental role takeover, which can mislead users and interfere with normal safety behavior or task routing.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The protocol mandates research on companies and key decision-makers using tools, but does not disclose that this may involve web access, data collection, or handling third-party information. In a negotiation-oriented skill, this context makes the issue more sensitive because users may be encouraged to gather personal or strategic information without clear consent, privacy, or compliance guardrails.

Content

No source excerpt is available for this finding.