Project Loop

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed long-running project orchestration helper, but it gives agents automatic cron-based persistence and retry behavior that users should review before installing.

Install only if you intentionally want an agent to manage long-running project state and recovery. Before using it, define the project scope and approval gates, inspect any created watchdog cron, confirm how to disable it, and do not allow retrying obfuscation-denied commands without explicit review.

SkillSpector

By NVIDIA

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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 (1)

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The skill instructs automatic creation and removal of watchdog cron jobs, which modifies scheduled system tasks without requiring explicit user awareness or approval at the point of action. Even though the watchdog is framed as bounded and safety-aware, persistence and autonomous scheduling expand the agent's execution surface and can lead to unintended background activity, privilege misuse, or hard-to-notice repeated actions.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal