Security audit

Calendar Reminder Tool Free

Security checks for vulnerabilities and agentic risk

Overview

This skill has a legitimate calendar-reminder purpose, but it combines automatic calendar access and Feishu messaging with overly broad, inconsistent instructions.

Review before installing. Only use this if you are comfortable with a nightly scheduled job reading your Outlook calendar and sending event details through Feishu. Configure the recipient carefully, verify the missing runtime script before enabling cron, and narrow or ignore the unrelated data-analysis trigger language.

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)

Intent-Code Divergence

Medium

Confidence: 92% confidence
Finding: The skill describes generic create/modify/delete/export-style operations and parameterized execution modes that do not match its stated purpose of scanning a calendar and sending reminders. This kind of scope inflation can cause an agent to over-grant capabilities or interpret unrelated user prompts as authorized actions, increasing the chance of unintended data access or modification.

Intent-Code Divergence

Medium

Confidence: 97% confidence
Finding: The trigger-condition section says the skill should be used for data analysis, reporting, statistical insights, and visualization, which directly contradicts the calendar-reminder function. In an agent setting, overly broad or incorrect trigger criteria can cause the skill to activate in unrelated contexts and operate on calendar or messaging integrations when the user did not intend that workflow.

Vague Triggers

High

Confidence: 96% confidence
Finding: The trigger description is both inaccurate and overly broad, making misrouting likely in an automated agent environment. If invoked for unrelated requests, the skill may read calendar data, use stored Outlook credentials, or send Feishu messages unexpectedly, creating privacy and operational risks.

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The capability coverage lists many broad keywords and fragmented scenarios without clear boundaries, which can bias an agent toward selecting this skill outside its legitimate scope. Ambiguous coverage claims are dangerous when the skill has access to calendar data and external messaging because accidental invocation can expose sensitive schedule information.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill processes calendar content using Outlook credentials and forwards schedule details through Feishu, but it does not clearly disclose what data is read, transmitted, retained, or who can receive it. This weak privacy notice can lead to unintentional disclosure of meeting titles, times, locations, and other potentially sensitive work information.

Static analysis

No suspicious patterns detected.