ClawHub
Back to skill

Security audit

Mp Caveman

Security checks for vulnerabilities and agentic risk

Overview

This skill only changes the assistant into a terse communication style, with no code execution, data access, or external connections.

Install only if you want the assistant to stay terse after activation. Use "stop caveman" or "normal mode" to turn it off, and avoid enabling it for tasks where fuller explanations matter.

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
92% confidence
Finding

The activation text uses broad phrases like requests for brevity, efficiency, or concise replies, which can match many normal user intents and silently switch the assistant into a persistent alternate behavior. That creates prompt-scope ambiguity and can override expected response style beyond the user's immediate request, increasing the chance of miscommunication or policy-relevant warnings being overly compressed.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The persistence rule says the mode remains active for every response once triggered and only turns off for two narrow opt-out phrases. This is dangerous because a loosely triggered style mode can persist across unrelated tasks, causing degraded clarity, hidden statefulness, and possible interference with safety-critical or high-stakes interactions despite the limited exception block.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.