Exec Local

• Outcome: Review
• Latest audit: May 29, 2026 at 9:30 PM UTC
• Version:
  1.0.0

Overview

This skill clearly aims to run shell commands, but that gives broad control over the container and possible SSH-managed hosts without clear guardrails.

Install only if you intentionally want an agent to run arbitrary shell commands in this container and possibly orchestrate a host through SSH. Use it in a tightly isolated environment, avoid untrusted prompts or command strings, and verify filesystem mounts, network reachability, and available credentials before enabling it.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The skill explicitly exposes arbitrary shell command execution inside a container and describes it as a general-purpose capability without any limitation, approval requirement, allowlist, or warning about the risks. Even if scoped to a container, this can enable destructive actions, data access, lateral movement through mounted secrets or network access, and privilege abuse depending on the runtime environment.

VirusTotal

65/65 vendors flagged this skill as clean.