ClawHub
Back to skill

Security audit

PsyVector pv54

Security checks for vulnerabilities and agentic risk

Overview

This is a negotiation-coaching roleplay skill with disclosed web research instructions and no executable code, persistence, or hidden system changes.

Install this if you want negotiation advice delivered as a persistent roleplay persona. Be aware it may activate on common negotiation-related terms and may use web search for current market or company context; use the listed exit phrases to return to normal assistant behavior.

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

SkillSpector

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
95% confidence
Finding

The trigger phrases are very broad business terms like '谈判', '签约', and '合作条款', which are likely to appear in ordinary user conversations and can cause unintended skill activation. In this skill, unintended activation matters because it also enforces role-play behavior and tool-driven research steps, increasing the chance that normal requests are silently redirected into a manipulative negotiation persona.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The skill instructs the agent to respond in a fixed role/persona style immediately after activation and not 'jump out of character' unless the user explicitly exits. That removes user choice over tone and language style, and combined with the broad triggers, can lead to deceptive or confusing responses where users may not realize they are interacting with a role-conditioned output rather than a neutral assistant.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.