ClawHub
Back to skill

Security audit

Google Drive (Composio)

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Google Drive integration with sensitive write and share powers, but I found no hidden or deceptive behavior.

Install only for a Composio-connected Google account you are comfortable letting an agent operate. Treat share, move, trash, and upload commands as sensitive actions, confirm the exact file IDs and recipients first, and keep local download/upload paths inside the intended workspace.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill documentation indicates capabilities to read environment variables, write files, and perform network operations, but it does not declare permissions explicitly. This is a real security weakness because users and platforms cannot accurately assess or constrain what the skill may access, especially since it handles API keys, downloads remote content, and writes data into the workspace.

Description-Behavior Mismatch

High
Confidence
99% confidence
Finding
The skill metadata says it is for browsing Google Drive folders and reading files, but the implementation exposes powerful write operations including create, update, move, copy, trash, and share. This capability mismatch is dangerous because an agent or reviewer may grant or approve the skill expecting read-only behavior, while the code can modify data and expand access to third parties.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.