ClawHub

eCRF Designer

Security checks across malware telemetry and agentic risk

Overview

This is a simple local CRF template generator that writes a user-named JSON file, with no evidence of hidden network access, credential use, persistence, or destructive behavior.

Install only if you are comfortable running a local Python script that writes a CRF JSON file. Run it in a workspace, avoid sensitive or system output paths, and review generated CRF specifications before using them in regulated clinical workflows.

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
89% confidence
Finding
The skill metadata advertises file-writing capability and local script execution behaviors, but it does not declare explicit permissions or enforcement boundaries for those actions. This creates a real security gap because agents or users may invoke the skill assuming a lower privilege profile than it actually needs, increasing the chance of unintended file modification in the workspace.

Missing User Warnings

Low
Confidence
80% confidence
Finding
The markdown states that Python/R scripts may execute locally and output files may be written, but it does not clearly warn users about the security implications of local execution, file creation, or possible overwrites. While the document is descriptive rather than executable, the missing user-facing warning can contribute to unsafe use and reduces informed consent around potentially risky behavior.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal