ClawHub
Back to skill

Security audit

Axioma Safe Cron Creator

Security checks across malware telemetry and agentic risk

Overview

This skill is meant to create cron jobs, but it gives an agent durable scheduling authority with broad triggers and incomplete safety boundaries.

Install only if you intentionally want an agent to create, verify, replace, and remove persistent user crontab entries. Review the exact crontab change and target script before running it, inspect the external helper scripts separately, and do not rely on the isolation claim beyond avoiding OpenClaw agent-session files.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The documentation makes an absolute safety claim ('no collision possible') for system cron execution, but the same skill allows arbitrary user-specified scripts and optional wrappers. Those scripts can still access shared files, session artifacts, network resources, or Telegram-related state, so the claim can mislead users into scheduling unsafe jobs under a false assumption of isolation.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The skill documents creation and removal of persistent cron entries and wrapper files without clearly warning that these operations change durable system scheduling state and may replace existing jobs. This can cause unintended persistence, accidental overwrite of legitimate tasks, or hard-to-notice recurring execution if invoked inappropriately.

Vague Triggers

Medium
Confidence
81% confidence
Finding
The trigger phrases are broad enough to match ordinary cron-related requests, increasing the chance the skill is invoked when the user did not intend to create or alter persistent system cron state. Because this skill performs system-level scheduling actions, accidental invocation materially raises the risk of unauthorized or surprising persistence.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.