Scada

Security checks across malware telemetry and agentic risk

Overview

This skill is a simple local log manager advertised as a SCADA industrial-control manager, so users should review it carefully before relying on it.

Install only if you want a small local entry/log helper. Do not use it for real SCADA monitoring, device control, safety workflows, or operational assurance. Avoid storing secrets or sensitive operational details, and be aware that removal is immediate and exports write local files.

SkillSpector

By NVIDIA

Vulnerability Patterns

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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Description-Behavior Mismatch

High

Confidence: 98% confidence
Finding: The skill is presented as a SCADA manager, but the implementation is only a local note/log utility. This mismatch is security-relevant because users may grant elevated trust, sensitive data, or operational access based on the claimed industrial-control purpose, while the code provides none of the expected safety, device-integration, or operational guarantees. In an agent ecosystem, deceptive capability claims can mislead operators into unsafe deployment decisions.

Intent-Code Divergence

High

Confidence: 96% confidence
Finding: The inline documentation and help output reinforce the false claim that this tool performs SCADA management, when it only manipulates local files. Misleading documentation can cause administrators or automated systems to invoke the skill in sensitive industrial contexts under false assumptions, increasing the risk of misuse, misplaced trust, and unsafe operational dependence.

Missing User Warnings

Medium

Confidence: 85% confidence
Finding: The documented `remove` command is destructive and provides no warning, confirmation behavior, or safety caveat. In an agent context, this increases the chance of accidental deletion of locally stored operational records, especially because the skill presents itself as SCADA-related and users may assume stronger safeguards around data handling.

Missing User Warnings

Medium

Confidence: 81% confidence
Finding: The `export` command writes data to a file without documented safeguards around destination path, overwrite behavior, or sensitivity of exported contents. In practice this can lead to unintended file overwrites or leakage of locally stored data into insecure locations, particularly in automation flows where file outputs may be handled non-interactively.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal