ClawHub
Back to skill

Security audit

Learning Evolution

Security checks across malware telemetry and agentic risk

Overview

This skill is a local template generator for reviewing skill usage, and its shell scripts are disclosed, scoped, and not designed to collect data, contact the network, or modify other skills.

Install only if you are comfortable running local shell scripts that create persistent Markdown reports. Review LEARNING_DATA_DIR if set, and avoid placing sensitive user feedback or identifiable logs into generated reports unless you intentionally want to retain that data.

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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill instructs users to execute local shell scripts, but the manifest does not declare permissions or clearly surface that shell execution is part of its operational model. This creates a transparency and trust-boundary issue: consumers may invoke code-capable behavior without explicit permission metadata, increasing the risk of unintended execution in environments that rely on declared capabilities for review or policy enforcement.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The README explicitly promotes tracking, analyzing, and evolving behavior based on skill usage and user interactions, but provides no privacy, consent, retention, or data-minimization guidance. In a skill focused on usage analytics, this omission increases the likelihood that operators collect or process potentially sensitive user interaction data without appropriate safeguards, creating privacy and compliance risk.

VirusTotal

38/38 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.