ClawHub

skll-scan

Security checks across malware telemetry and agentic risk

Overview

This is a user-run local scanner, but it overstates its threat-intelligence protection and could give users more confidence than the code supports.

Install only if you treat this as a lightweight local static scanner, not proof that a skill is safe. Avoid scanning code containing secrets unless you control or delete the generated /tmp report, and do not enable or copy third-party lookup examples for internal, customer, or confidential indicators without approval.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
93% confidence
Finding
The skill documentation and described usage indicate capabilities for reading and writing files, making network requests to threat-intelligence services, and invoking shell commands, yet the skill declares no permissions. That mismatch is dangerous because it can cause users or a host platform to trust and install a skill without understanding its actual access needs, weakening least-privilege controls and auditability.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The example immediately demonstrates sending a host/domain to a third-party threat intelligence service without a nearby explicit warning that submitted indicators may reveal sensitive internal infrastructure, customer domains, or investigation targets. In a security-scanning skill, users are especially likely to paste suspicious or private indicators, so omission of an in-context privacy warning can cause unintended data disclosure.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal