Security audit

Gmail IMAP

Security checks across malware telemetry and agentic risk

Overview

This Gmail skill is transparent about accessing Gmail, but a UID-handling mismatch could make it read, move, or trash the wrong email.

Install only for trusted agents and preferably use a dedicated, revocable Google App Password. Be especially careful with read, move, and trash commands until UID handling is fixed or manually verified.

SkillSpector

By NVIDIA

Vulnerability Patterns

MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
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 (2)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 95% confidence
Finding: The skill clearly requires sensitive environment variables and performs networked Gmail operations, but the manifest does not declare corresponding permissions. This creates a policy and transparency gap: an agent or platform may invoke a skill with mailbox-reading, email-sending, and deletion capabilities without explicit permission gating, increasing the risk of unintended data access or destructive actions.

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: The README explicitly documents destructive operations like trashing and moving email without clearly warning that these actions can permanently affect a user's mailbox state or cause unintended data loss. In an agent skill that grants mailbox access, normalizing destructive commands without caution increases the chance that an agent or user invokes them unsafely, especially in automated workflows.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.