Crowdstrike

Security checks across malware telemetry and agentic risk

Overview

This appears to be a legitimate CrowdStrike integration, but it delegates broad authenticated control over a sensitive security platform without enough scoping or confirmation guidance.

Install only if you trust Membrane and intend to let an agent work with CrowdStrike. Use a dedicated least-privilege CrowdStrike account, consider pinning the Membrane CLI version, and require manual approval before any POST, PUT, PATCH, DELETE, RTR, or other operational response action.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill explicitly documents a generic proxy request mechanism supporting POST, PUT, PATCH, and DELETE against the CrowdStrike API without guardrails, confirmation requirements, or warnings about destructive effects. In a security-platform context, this could enable unintended state changes such as modifying detections, sessions, or other sensitive resources if an agent uses raw requests instead of safer, scoped actions.

VirusTotal

54/54 vendors flagged this skill as clean.

View on VirusTotal