ClawHub
Back to skill

Security audit

Mlscp

Security checks across malware telemetry and agentic risk

Overview

This skill is a documentation-only reference for a compact command protocol, and its risky file-operation terms are syntax examples rather than executable behavior.

Install this as a protocol-reference skill only if you want MLSCP notation. Do not allow compact commands from other agents to directly read, modify, or delete files without expanding them into plain language, validating the target paths, and getting user approval. Separately inspect any referenced scripts or Python package before running them.

SkillSpector

By NVIDIA
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
Confidence
91% confidence
Finding
The manifest description is overly broad: it activates on generic themes like communicating with other agents, parsing compressed commands, or token-efficient instructions. That can cause the skill to trigger in ordinary conversations and route users into a command protocol that may encode file or variable operations, increasing the chance of unintended capability invocation or misuse.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The documentation exposes destructive operations like file delete and variable delete as normal commands without any safety warning, confirmation requirement, or boundary guidance. In the context of an agent skill meant for inter-agent command compression, this makes dangerous write/delete actions easier to invoke or relay, especially if another component treats parsed commands as trusted instructions.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.