Security audit

Cron Optimizer

Security checks across malware telemetry and agentic risk

Overview

This cron cleanup skill modifies scheduled jobs automatically and can recreate a watchdog job without clear user approval or tight scoping.

Install only if you intentionally want this skill to mutate OpenClaw cron state and preserve the evolver watchdog. Review the cron jobs it may remove first, and prefer adding a dry-run/confirmation path and a strict allowlist before running it on a real system.

SkillSpector

By NVIDIA

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (7)

Description-Behavior Mismatch

High

Confidence: 97% confidence
Finding: The skill is presented as a cron cleanup utility, but it also restores a specific watchdog job when missing. That hidden persistence behavior goes beyond stated functionality and can re-establish agent infrastructure after an operator intentionally removed it, undermining user control and creating a persistence mechanism.

Context-Inappropriate Capability

High

Confidence: 98% confidence
Finding: Launching `node skills/feishu-evolver-wrapper/lifecycle.js ensure` is unrelated to stale cron cleanup and invokes external lifecycle management to restore infrastructure. This creates covert cross-skill coupling and enables reconstitution of scheduled execution even when the watchdog was deliberately removed, which is a strong persistence signal.

Intent-Code Divergence

Medium

Confidence: 95% confidence
Finding: The inline documentation claims the skill only cleans stale and redundant cron jobs, but the implementation also recreates a missing watchdog. This mismatch is dangerous because it conceals persistence behavior from reviewers and operators, increasing the chance the skill is trusted and deployed without understanding its real effects.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill advertises removal of cron jobs and cleanup of job definitions without warning the user that these are destructive system changes. In a system-management context, silently deleting scheduled tasks can disable maintenance, security updates, backups, or monitoring, making accidental misuse materially harmful.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The script automatically removes cron jobs based on heuristics without confirmation, dry-run mode, or strong allowlisting. In a system-management context, mistaken classification can delete legitimate jobs and cause operational disruption or loss of scheduled security/maintenance tasks.

Missing User Warnings

Low

Confidence: 92% confidence
Finding: The script silently recreates a watchdog by spawning another Node process, with only a warning log and no explicit authorization. In context, this is more dangerous because the skill is nominally a cleanup tool, so users would not expect it to restore persistent scheduled execution behind the scenes.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The package description explicitly states that the skill 'remov[es] stale, disabled, or redundant' cron entries, which implies modification of system scheduler configuration. In a skill package, omitting any warning, consent requirement, or scope limitation for such behavior is risky because users may invoke it without understanding that it can delete scheduled tasks and affect system operation or persistence mechanisms.

VirusTotal

VirusTotal engine telemetry is currently stale for this artifact.

View on VirusTotal