ClawHub
Back to skill

Security audit

Qcsd Quality Gates

Security checks across malware telemetry and agentic risk

Overview

This is a code-quality helper, but it gives broad automatic repair instructions without clear user approval and its runtime reports success without actually checking the project.

Review before installing. Treat this skill as a checklist, not a reliable automated gate, unless the implementation is completed. Use version control and require explicit approval before any dependency changes, generated files, or broad auto-fixes.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • 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 (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger conditions are written as mandatory and broad (e.g., every new project, every code output, immediate forced execution on user request), which can cause the skill to activate in routine interactions without clear user consent or bounded scope. In an agent setting, this increases the chance of overreach, unintended workflow hijacking, and automatic modification behavior being applied when not appropriate.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger conditions are overly broad and mandatory (e.g., every new project, every code output, immediate forced execution on user quality requests), which can cause the skill to activate during ordinary development interactions without clear user intent. In a skill that can auto-fix code and files, unintended invocation increases the chance of unauthorized or surprising changes to a user's workspace.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill advertises automatic dependency changes, code fixes, startup-failure remediation, and file creation, but does not state any confirmation, preview, rollback, or risk disclosure requirements. Because these actions directly modify the project, an unintended or incorrect repair could break builds, alter behavior, or introduce unsafe changes without the user's informed consent.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal