Google Meet

Security checks across malware telemetry and agentic risk

Overview

This Google Meet skill is mostly coherent, but it needs review because it gives an agent authenticated access to sensitive meeting records and a broad direct API proxy without clear safety guardrails.

Install only if you are comfortable letting an agent use your Membrane-authenticated Google Meet connection. Prefer the listed pre-built actions, review exact endpoints and payloads for proxy requests, and require explicit approval before retrieving transcripts or recordings or making POST, PUT, PATCH, or DELETE requests.

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 (2)

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The skill explicitly supports access to recordings and transcripts, which can contain highly sensitive meeting content, but the documentation provides no user-facing warning or consent guidance. In an agent setting, this increases the risk of users or downstream systems invoking the skill without appreciating that confidential communications may be retrieved or exposed.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The documented proxy feature allows direct requests to the Google Meet API with authenticated access, but there is no warning that arbitrary requests may transmit or retrieve sensitive meeting metadata or content over the network. This expands capability beyond curated actions and can enable broader data access with fewer guardrails, especially in agent-driven workflows.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal