ClawHub
Back to skill

Security audit

Manage Agent Skills

Security checks across malware telemetry and agentic risk

Overview

This is an on-demand skill manager that can change agent skill settings, but the behavior is disclosed, scoped, and backed up.

Install this only if you want an agent to manage skill enablement across local agent platforms. Use dry-run first, review the reported backup path, and be careful with presets or the all selector because they can change which skills are active in future agent sessions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill instructs the agent to invoke a local Python script, read reference files, inspect installed skills, and mutate platform configuration files, which implies shell, file read, file write, and environment access. Because these capabilities are not explicitly declared, users and enforcement layers may underestimate the privilege level of the skill, increasing the chance of unsafe approval or execution in sensitive environments.

VirusTotal

54/54 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.