Security audit

Cron Scheduler Setup

Security checks across malware telemetry and agentic risk

Overview

This is a cron scheduler skill, but it can create persistent background agent jobs and send scheduled results to external channels without enough scoping or privacy guidance.

Review any scheduled job before enabling it, especially isolated-session jobs and Telegram, WhatsApp, or webhook delivery. Confirm the exact schedule, message contents, destination, and how to list or delete the job, and avoid scheduling summaries of private email, calendar, account, or business data unless you intend that data to be sent repeatedly.

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 (3)

Intent-Code Divergence

Medium

Confidence: 91% confidence
Finding: The description contradicts the actual purpose of the skill by claiming it is for code generation/programming assistance while the rest of the document describes a cron scheduler. This kind of scope confusion can cause the agent to invoke the skill in inappropriate contexts, increasing the chance of unintended command execution or scheduling actions under false assumptions about what the skill does.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The trigger condition is broad enough to match many generic automation or scheduling-adjacent requests, which can cause over-selection of a skill that has exec/write capabilities. In an agent environment, overly broad routing increases the risk of the skill being invoked for tasks the user did not intend, potentially leading to unauthorized task creation, file changes, or command execution.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The documentation includes examples that deliver scheduled outputs to external channels such as Telegram and WhatsApp, but it does not clearly warn users that data may leave the local environment. In a scheduler skill, this is more dangerous because jobs may run later and repeatedly, causing ongoing exfiltration of message contents, summaries, or other sensitive information without contemporaneous user awareness.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.