Security audit

日历同步工具基础版

Security checks across malware telemetry and agentic risk

Overview

The skill is a calendar-sync assistant, but it under-discloses cloud synchronization and credential handling for actions that could change external calendars.

Review this before installing. It may read configuration and credentials, use command execution, contact external calendar providers, and create or synchronize events. Only use it if you are comfortable with those actions, confirm timezone and target platforms before any event creation, and avoid running the broad environment-variable check.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Description-Behavior Mismatch

Medium

Confidence: 93% confidence
Finding: The skill instructs the agent to enumerate environment variables matching API/KEY/TOKEN/SECRET during configuration verification, which unnecessarily exposes the presence and names of sensitive secrets to the skill workflow. In a skill that also advertises external API/network usage, this creates an avoidable secret-discovery surface and directly conflicts with the claim that the tool is purely local/private.

Intent-Code Divergence

High

Confidence: 97% confidence
Finding: The skill states that all FREE-version data is stored locally and not uploaded to the cloud, while elsewhere it explicitly supports syncing with Google, Apple, and Outlook and notes that some functions require external API access. This contradiction can mislead users into authorizing actions under false privacy assumptions, causing unintended disclosure or modification of calendar data on third-party platforms.

Vague Triggers

High

Confidence: 88% confidence
Finding: The trigger condition says to use the skill for data analysis, report generation, statistical insight, and visualization, which is unrelated to the stated calendar-management purpose. Overbroad and mismatched trigger logic increases the chance an agent invokes this skill in inappropriate contexts, leading to unintended command execution, wrong actions on user data, or unsafe cross-tool behavior.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill describes cross-platform synchronization and event operations but does not clearly warn users that actions may create, modify, or propagate calendar data across external services. Without an upfront mutation warning and confirmation boundary, users and agents may treat the skill as read-only or local-only, causing accidental data changes on connected accounts.

Natural-Language Policy Violations

Medium

Confidence: 84% confidence
Finding: The sample configuration hard-codes the timezone to Asia/Shanghai without documenting user opt-in or safe defaults, which can cause event creation and reminders to be scheduled at incorrect times. In a calendar tool, silent locale/timezone assumptions are especially risky because they directly affect real-world meetings, reminders, and cross-platform synchronization behavior.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.