Crowdpower

Security checks across malware telemetry and agentic risk

Overview

This skill is transparent about using Membrane for CrowdPower, but it gives an agent broad authenticated power to change business records through raw API requests without clear guardrails.

Install only if you are comfortable letting the agent operate your CrowdPower account through Membrane. Use a least-privilege account where possible, review exact action IDs, endpoints, and JSON inputs before execution, and require confirmation for bulk, charge-related, POST, PUT, PATCH, or DELETE requests.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The skill is presented as a bounded CrowdPower integration, but it explicitly exposes a generic authenticated proxy for arbitrary API paths. That widens capability beyond the declared scope and can enable unintended high-impact operations, sensitive data access, or destructive requests if an agent uses the proxy instead of constrained actions.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal