育儿助手

• Outcome: Review
• Latest audit: May 29, 2026 at 9:32 PM UTC
• Version:
  1.1.0

Overview

The skill is a coherent baby-care helper, but its vaccine feature can automatically change calendars and create scheduled reminders without clear consent or cleanup controls.

Install only if you are comfortable reviewing the vaccine feature carefully. Before using it, require the agent to show the next-dose calculation, calendar title/date/notes, and reminder time, then ask for explicit approval before it writes anything. Also verify that the external apple-calendar skill is installed and trusted, and know where cron reminders are created so you can remove them.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The README presents the skill as providing health, fever-handling, vaccine scheduling, and care guidance for infants without any disclaimer that the content is informational only and not a substitute for pediatric or emergency care. In a baby-care context, users may rely on the tool during urgent situations, increasing the risk of delayed treatment, unsafe self-management, or overconfidence in AI-generated medical advice.

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The trigger list is very broad and overlaps with common family, health, and child-related conversation, which can cause the skill to activate outside the user's clear intent. In this skill, accidental activation is more dangerous because the skill is also configured to perform downstream actions such as vaccine scheduling and reminders, increasing the chance of unintended side effects.

Missing User Warnings

High

• Confidence: 98% confidence
• Finding: The skill description states that when a user mentions a vaccine, the system will automatically calculate the next dose and add calendar/reminder entries without clearly disclosing that it will modify user systems. This creates an unauthorized action risk and, in a healthcare-adjacent context, can also lead to incorrect or unwanted scheduling of medically significant reminders.

Missing User Warnings

High

• Confidence: 99% confidence
• Finding: The workflow explicitly instructs the agent to call a calendar-writing script and create a cron-based reminder as mandatory steps, yet it does not require user approval before making persistent changes. This is especially risky because the payload includes a directive to bypass normal messaging behavior ('禁止调用message工具'), which reduces transparency and makes silent side effects more likely.

VirusTotal

66/66 vendors flagged this skill as clean.