ClawHub

Cloudmersive

Security checks across malware telemetry and agentic risk

Overview

This appears to be a real Cloudmersive integration, but it is broad enough to send sensitive files, text, URLs, and documents to external services without a clear consent boundary.

Install only if you specifically want to use Cloudmersive through Membrane. Before running actions or proxy requests, review exactly what file, URL, text, email address, or document will be sent externally, avoid sensitive or regulated data unless approved, and prefer curated actions over raw proxy requests.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
91% confidence
Finding
The manifest description says the skill is for managing Cloudmersive data, records, and workflows, but the actual documented capabilities are malware scanning, OCR, barcode/image processing, URL validation, and document conversion. This mismatch can cause an agent or user to invoke the skill under false assumptions and unintentionally send sensitive content to third-party services outside the expected scope.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger 'Use when the user wants to interact with Cloudmersive data' is overly broad and underspecified. Broad invocation criteria increase the chance that an agent will select the skill for loosely related tasks and transmit files, text, URLs, or documents to Cloudmersive without clear user intent or informed consent.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill encourages running actions and proxy requests but does not warn that user-provided files, text, URLs, emails, or documents may be sent to external Cloudmersive infrastructure. In this context, the omission is significant because many listed actions inherently involve uploading potentially sensitive content, creating privacy, compliance, and data handling risks if the user is not informed.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal