Baby (Tracker, Feeding, Sleep, Triage, Visit Prep)

• Outcome: Pass
• Latest audit: May 29, 2026, 8:22 PM
• Version:
  1.0.0

Overview

This is a local, instruction-only baby-care tracking skill that clearly discloses sensitive local note storage and does not show hidden code, network use, or credentials.

Install only if you are comfortable keeping baby-care and pediatric notes in local ~/baby/ files. Confirm writes before saving, store only the minimum useful details, limit access to trusted caregivers, and periodically delete or archive old health notes.

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 (2)

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The setup directs the skill to activate on a very broad set of common baby-care topics and to clarify this behavior within the first exchanges. In a family or parenting assistant context, these subjects arise frequently, so the skill could be invoked unexpectedly, causing unnecessary data collection, intrusive prompting, or safety-sensitive guidance to appear when the user did not intend to use the skill.

Missing User Warnings

Medium

• Confidence: 85% confidence
• Finding: The template directs users to store post-visit instructions in a persistent "handoff and memory" system, which can include sensitive pediatric health information such as symptoms, medications, and follow-up plans. Without any guidance on minimizing, securing, or limiting retention of this data, the skill encourages handling health-related information in a way that may expose private medical details to unintended storage, sharing, or later retrieval contexts.

VirusTotal

66/66 vendors flagged this skill as clean.