ClawHub
Back to skill

Security audit

Google 日历基础版

Security checks across malware telemetry and agentic risk

Overview

This calendar skill is mostly coherent, but it should be reviewed because it understates cloud data flow and allows destructive calendar changes without a clear confirmation requirement.

Review this skill before installing. Only use it with calendar accounts where command-line Google Calendar or CalDAV access is acceptable, avoid providing callback URLs unless you trust the destination, and require the agent to show matched events and get explicit confirmation before deleting anything.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
83% confidence
Finding
The skill is presented as a local Google Calendar utility, but it also advertises a callback_url for asynchronous completion notices, which introduces outbound network communication not clearly aligned with the local-only framing. This can mislead users and agents into sending task metadata or calendar-related results to arbitrary external endpoints, increasing privacy and exfiltration risk.

Intent-Code Divergence

High
Confidence
97% confidence
Finding
The documentation claims all data is stored locally and not uploaded to the cloud, but the skill explicitly depends on Google Calendar APIs and optionally CalDAV servers, both of which involve transmitting calendar data to external services. This is a material security and privacy misrepresentation that can cause users to expose sensitive scheduling information under false assumptions about data locality.

Vague Triggers

High
Confidence
89% confidence
Finding
The trigger conditions are overly broad and inconsistent with the actual calendar-management capabilities, including references to data analysis and visualization that the FREE tool does not actually implement. In an agentic environment, vague trigger logic can cause the skill to be invoked in unintended contexts, increasing the chance of unnecessary command execution, data access, or destructive calendar operations.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill includes a delete-by-keyword capability but does not require an explicit confirmation step, dry run, or clear warning about destructive effects. In an AI-agent setting, ambiguous natural-language matching could cause accidental deletion of one or more calendar events, leading to integrity loss and operational disruption.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.