Security audit

Ai Intelligent Report Generator

Security checks across malware telemetry and agentic risk

Overview

This looks like a legitimate report-generation skill, but it asks users to run unscanned external code and describes automatic email sending for potentially sensitive reports without clear controls.

Review the referenced GitHub repository, dependencies, and app.py before running the install commands. Install only if you can verify how email credentials are stored, who receives reports, whether each send is approved, and how scheduled sending and retained report versions can be disabled or deleted.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The skill description is overly broad and lacks clear activation boundaries, triggers, or constraints on what data it may analyze and what reports it may generate. In an agentic environment, ambiguous scope increases the risk of unintended processing of sensitive data or execution in contexts the user did not explicitly authorize.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The documentation advertises automatic sending and email integration without warning that generated reports or analyzed data may be transmitted externally. This creates a material risk of unintentional data exfiltration, especially if reports contain financial, project, or other sensitive business information and outbound delivery occurs automatically.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.