universal-shell-deployer

• Outcome: Review
• Latest audit: May 29, 2026 at 9:35 PM UTC
• Version:
  1.0.0

Overview

This skill is transparent about running deployment commands, but it needs review because it can change local or remote systems and includes unsafe default administrative credentials.

Install only if you are comfortable reviewing and approving shell commands that can modify services on the selected machine or remote host. Replace the MinIO placeholder password before any use, avoid root or Administrator remote accounts where possible, keep example remote nodes disabled until intentionally configured, and do not store secrets or sensitive command output in config.json.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: The skill advertises a very broad capability to install, configure, and operate infrastructure across local and remote systems, which makes it likely to be invoked for many generic setup requests without sufficient narrowing. In practice, an over-broad trigger on a command-executing skill increases the chance that an agent will select a high-privilege shell automation path for ambiguous requests, leading to unintended command execution on local or remote hosts.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The configuration contains a remote SSH example using the username "root" and also defines default MinIO credentials ("minioadmin" / "change-me"). In a skill designed to execute infrastructure setup on local or remote hosts, embedded privileged accounts and default secrets materially increase the chance of unsafe deployment, credential reuse, accidental exposure, or immediate compromise if the generated environment is network-accessible.

Vague Triggers

Low

• Confidence: 76% confidence
• Finding: This manifest does not clearly constrain when the skill should activate or what user intent is required before it can run command-execution workflows. Because the skill is explicitly capable of installing software and operating on remote hosts over SSH/PowerShell remoting, ambiguous trigger scope raises the risk of unintended invocation, overbroad use, or execution in contexts the user did not clearly authorize.

VirusTotal

66/66 vendors flagged this skill as clean.