ClawHub
Back to skill

Security audit

Wol Sleep Pc

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small local-network wake/sleep packet sender with some documentation gaps, but no hidden persistence, exfiltration, or deceptive behavior was found.

Install only if you control the target PC and LAN. Configure the MAC, sleep_mac, broadcast address, and port carefully; the skill does not verify device status, and the sleep helper expects you to supply the inverted sleep MAC yourself. Treat the optional config file as local device information and keep it private.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (6)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding
The skill advertises support for Sleep-on-LAN, inverted MACs, target IP handling, and status checking, while the analyzed behavior indicates only broadcast packet sending with a provided MAC-like value. In an agent setting, this discrepancy can mislead automation into performing broader broadcast-based network operations than expected, creating reliability and misuse risks on the local network.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding
The skill advertises support for Sleep-on-LAN, inverted MACs, target IP handling, and status checking, while the analyzed behavior indicates only broadcast packet sending with a provided MAC-like value. In an agent setting, this discrepancy can mislead automation into performing broader broadcast-based network operations than expected, creating reliability and misuse risks on the local network.

Intent-Code Divergence

High
Confidence
98% confidence
Finding
The documentation states this script sends a Sleep-on-LAN packet 'using inverted MAC', which implies the MAC should be bitwise inverted before packet construction. However, the implementation only strips separators and repeats the supplied value directly in a standard Wake-on-LAN-style payload, so the documented intent contradicts actual behavior.

Session Persistence

Medium
Category
Rogue Agent
Content
python3 scripts/send_sleep.py --mac 99:90:CA:FE:4B:24 --broadcast 192.168.1.255

Config file
Create ~/.config/wol-sleep-pc/config.json with keys:
{
  "mac": "24:4B:FE:CA:90:99",
  "sleep_mac": "99:90:CA:FE:4B:24",
Confidence
60% confidence
Finding
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Lp3

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding
Without declared permissions the skill's intent is opaque and cannot be validated.

Description-Behavior Mismatch

Medium
Confidence
83% confidence
Finding
The manifest advertises use when the user asks to 'wake, check, or put the PC to sleep', implying some capability to inspect or verify PC state. This script only parses arguments and sends a UDP packet; there is no code here to query reachability, power state, or any other status indicator.

Static analysis

No suspicious patterns detected.