ClawHub
Back to skill

Security audit

Cloudflare MCP

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly coherent for Cloudflare MCP use, but it exposes a broad execute pathway that can call Cloudflare API endpoints without clearly separating read-only, write, and destructive operations.

Review this skill before installing if the connected Cloudflare account has production or administrative access. It should be used only when you are comfortable with OOMOL-mediated Cloudflare MCP actions, and any use of `execute` should be treated as potentially capable of changing Cloudflare resources unless the live schema and payload prove it is read-only.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The trigger text instructs the agent to use this skill for ANY Cloudflare MCP request and, more broadly, whenever a task involves Cloudflare MCP, instead of calling the API directly. That broad routing can cause unintended invocation for loosely related Cloudflare tasks, increasing the chance the agent executes connector-backed actions in contexts where a narrower skill or direct, more controlled workflow would be safer or more appropriate.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.