ClawHub
Back to skill

Security audit

Automation Recipe Pack

Security checks across malware telemetry and agentic risk

Overview

This is a broad automation helper that appears to request command-capable authority without clearly scoping when commands may run or what they may change.

Review this skill before installing. Use it only for clearly requested automation tasks, ask the agent to show any planned commands first, and avoid using it in sensitive directories or with secrets until the publisher narrows the triggers and documents command side effects.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The skill's activation language is extremely broad, describing generic productivity, automation, batch processing, and workflow optimization use cases that overlap with many normal user requests. Because the skill also advertises exec capability, this broad routing can cause the agent to invoke a command-capable skill in situations where users did not explicitly request system-level actions, increasing the chance of unintended command execution or privilege misuse.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The capability coverage section lists vague natural-language phrases such as efficiency improvement and workflow optimization as covered scenarios without defining boundaries or safeguards. In a skill with exec enabled, these vague triggers increase the likelihood of over-selection and unsafe interpretation of ordinary requests as permission to automate or execute commands.

Missing User Warnings

High
Confidence
97% confidence
Finding
The skill metadata exposes exec capability, but the descriptive sections do not warn users that commands may be executed or that local environment state may be changed. This is particularly dangerous because the surrounding text frames the skill as a generic beginner-friendly automation helper, which can lower user caution and lead to unsafe invocation of shell actions without informed consent.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The usage workflow instructs users to run the skill and troubleshoot execution issues but omits any warning that command execution may have side effects on the host system. By presenting execution as routine and providing operational steps without a safety checkpoint, the skill increases the risk of users approving or triggering actions that alter files, network state, or system configuration unintentionally.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.