Skills Cli

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:19 PM UTC
• Version:
  1.0.2

Overview

This is a documentation-only helper for ClawHub skill management; its broad update and sync commands need care but fit the stated purpose.

Install is reasonable as a command-help skill. Before running its suggested bulk commands, check the active ClawHub account, the current directory, which skill folders are included, and whether any files contain secrets or private material.

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 (4)

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The activation guidance is broad enough to match generic requests about 'managing skills' or 'skills cli', which can cause this skill to trigger in situations where a more specific skill or direct user clarification would be safer. Over-broad routing increases the chance of unintended command suggestions, especially for install, update, publish, or sync operations that affect local or remote state.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The description defines a very broad scope spanning search, install, update, publish, and organization, without clear boundaries for when this orchestrator should defer or ask for confirmation. In an agent environment, such vague scope can lead to accidental invocation and inappropriate guidance for state-changing operations.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The skill recommends publish and sync commands that send local skill contents to a remote service but does not warn that local files, metadata, or embedded secrets may be uploaded externally. In this context, the omission is dangerous because users may treat sync/backup as harmless local maintenance and unknowingly exfiltrate sensitive content.

Missing User Warnings

Low

• Confidence: 88% confidence
• Finding: The example promotes `clawhub update --all` without highlighting that it performs bulk changes across all installed skills, which may introduce breaking changes or unexpected behavior. While not inherently malicious, omitting the caution reduces informed consent for a broad state-changing operation.

VirusTotal

49/49 vendors flagged this skill as clean.