ClawHub
Back to skill

Security audit

Sshot

Security checks across malware telemetry and agentic risk

Overview

This skill is a simple screenshot helper, but it runs an unbundled local PowerShell script with broad screen-capture access and too little user control.

Review before installing. Only use it on a Windows node you control, inspect the referenced sshot.ps1 file first, and close or hide sensitive windows before running /sshot. Prefer a version that bundles the script, avoids ExecutionPolicy Bypass where possible, and asks for confirmation before full-screen capture.

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 (3)

Missing User Warnings

Medium
Confidence
96% confidence
Finding
This skill triggers a full-screen screenshot on a Windows node and returns the screenshot file path, but it does not warn the user that the capture may include passwords, private messages, tokens, customer data, or other sensitive on-screen content. In the context of an agent skill, this omission is dangerous because users may invoke it without informed consent, and the command uses PowerShell with ExecutionPolicy Bypass, which increases the sensitivity of the action even if the core issue here is inadequate disclosure.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The skill is bound to a simple `/sshot` trigger with no scope restrictions, confirmation step, or exclusion criteria, making it easy to invoke screenshot capture in situations the user may not fully anticipate. Because the action captures the full screen on a Windows node, accidental or overly broad invocation can expose sensitive on-screen data such as credentials, messages, or confidential documents.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill performs full-screen screenshot capture and returns the output path without any explicit privacy warning, consent language, or notice about the sensitivity of the collected data. Full-screen capture can include passwords, tokens, personal communications, internal dashboards, or regulated data, so the lack of an upfront warning materially increases the risk of unintended data collection and disclosure.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.