Security audit

Py Data Analyzer Free

Security checks across malware telemetry and agentic risk

Overview

This skill is a normal Python data-analysis helper, but it asks for broad execution and file authority and includes export/callback paths without clear user controls.

Review this before installing if you will use it with sensitive datasets. Only allow execution, file writing, exports, or callback URLs when you explicitly request them and understand where results will be saved or sent.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Description-Behavior Mismatch

Medium

Confidence: 93% confidence
Finding: The skill is presented as a knowledge-base helper, but it declares powerful write and exec capabilities and also describes create/export operations. That mismatch can cause an agent or user to trust the skill as read-only guidance while it is actually capable of executing code and modifying files, increasing the risk of unintended command execution or data writes.

Intent-Code Divergence

Medium

Confidence: 89% confidence
Finding: The documentation claims the skill is a pure knowledge-base tool needing no extra setup, but later states it is MD+EXEC and that some functions require command execution. This inconsistency can mislead users and orchestrators into granting trust or invoking the skill in contexts where code execution is not expected.

Vague Triggers

Medium

Confidence: 81% confidence
Finding: The trigger condition is broad enough to match many generic data-help requests, which increases the chance the skill is selected in situations beyond its intended scope. Because the skill also advertises exec and write capabilities, overbroad activation raises the risk of unnecessary code execution or file/output actions during routine conversations.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The skill documents create/query/export operations without warning that export may write files or emit sensitive data. In a data-analysis context, exports can contain confidential business or research data, so omitting disclosure and confirmation steps increases the risk of unintended data leakage or file modification.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The documented callback_url allows results to be sent to an external endpoint without any privacy, trust-boundary, or network-transmission warning. In a data-analysis skill, results may contain proprietary or personal data, so silent callbacks can enable unintended exfiltration to attacker-controlled URLs.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.