ClawHub

Mobivate

Security checks across malware telemetry and agentic risk

Overview

This appears to be a real Mobivate/Membrane integration, but it needs review because its description is inconsistent and it allows broad authenticated API actions.

Install only if you intend to use Membrane as an intermediary for Mobivate. Verify the CLI source and version, connect the least-privileged Mobivate account available, and require explicit approval before sending SMS messages, changing or deleting contacts/lists/templates, or accessing account/API-key settings.

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 (4)

Description-Behavior Mismatch

High
Confidence
95% confidence
Finding
The manifest advertises a CRM-style integration for managing Persons, Organizations, Deals, Leads, Projects, and Activities, but the body documents a marketing-automation/SMS service and permits arbitrary API proxying. This mismatch can cause an agent or user to invoke the skill under false assumptions, leading to over-broad access, unintended actions, or data being sent to the wrong service context.

Intent-Code Divergence

High
Confidence
94% confidence
Finding
The skill documentation is internally contradictory about what product and data model it supports, which undermines safe tool selection and reliable operator understanding. In agentic systems, ambiguous capability claims are dangerous because they can trigger the wrong skill for a user request and cause unintended reads, writes, or credentialed requests against an external API.

Vague Triggers

Medium
Confidence
77% confidence
Finding
The activation description is broad enough that an orchestrator could select this skill for many vague requests involving 'Mobivate data,' even when a narrower or safer tool would be more appropriate. Because the skill supports authenticated external actions, over-selection increases the chance of unnecessary data exposure or unintended side effects.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The documentation explicitly enables direct API proxying but does not warn that arbitrary paths, methods, headers, and bodies may transmit sensitive data or trigger state-changing operations. In a skill intended for agent use, that omission raises the risk of silent exfiltration, unsafe write operations, and user surprise about what data is being sent externally.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal