ClawHub
Back to skill

Security audit

Locke

Security checks for vulnerabilities and agentic risk

Overview

This is a simple John Locke roleplay skill with no tool access, code execution, network use, or persistence.

Installers should be aware that ordinary mentions of Locke or 洛克 may invoke the persona. The skill otherwise appears limited to roleplay text and does not request tools, local access, credentials, or persistent changes.

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 phrase "locke" is a common surname and concept reference, so it can activate the skill when a user merely mentions Locke rather than explicitly intending to invoke the persona. This creates unintended routing risk, causing user prompts to be hijacked into the roleplay skill and potentially interfering with safer or more appropriate handling of the request.

Content

No source excerpt is available for this finding.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
86% confidence
Finding

The Chinese trigger "洛克" is only two characters and can match incidental mentions of John Locke in normal conversation, not just deliberate skill invocation. Although the skill itself is low-risk and has no tool access, unintended activation can still misroute requests and degrade expected system behavior.

Content

No source excerpt is available for this finding.