Security audit

Auto Data Analyzer

Security checks across malware telemetry and agentic risk

Overview

This is a simple documentation-only data analysis skill whose local report generation is expected and disclosed, with ordinary privacy and file-output caution needed.

Install the Python packages in a virtual environment, and only run the examples on datasets you intentionally want analyzed. Generated reports and charts may contain sensitive values from the source data, so keep outputs in a controlled folder and avoid sharing them unless the data is safe to disclose.

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 (2)

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The skill description is very broad and describes generic automated data analysis without clear triggering boundaries, input constraints, or user-intent checks. In an agent environment, this can cause over-invocation on ordinary data-related prompts and lead to unintended processing of user files or generation of derived artifacts without sufficiently explicit consent.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The skill documents output artifacts such as HTML reports, PNG charts, and CSV results, but it does not prominently warn that execution writes files to disk. In agentic contexts, silent file creation can surprise users, overwrite existing files, expose sensitive derived data in persistent artifacts, or create unsafe HTML outputs that are later opened or shared.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.