Security audit

happy-brushing-hero

Security checks across malware telemetry and agentic risk

Overview

This skill is a local child brushing timer and reward tracker with disclosed local data storage, but parents should be aware it can persist brushing records and may activate on broad everyday phrases.

Install only if you are comfortable with local storage of a child's brushing name, schedule, streaks, and sticker progress under ~/.bookshelf-plus/kids. Use explicit commands such as starting the brushing timer, avoid broad automatic triggers, and periodically delete the local data files if you do not want long-term records.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 91% confidence
Finding: The skill advertises executable Python scripts that read and write user data under the home directory and can invoke shell-like behavior via subprocess or optional system speech integration, yet no permissions are declared. This creates a trust and review gap: an agent or user may run the skill without understanding that it can modify files and execute local commands.

Vague Triggers

High

Confidence: 95% confidence
Finding: The trigger keywords are broad everyday phrases such as '牙齒' and '幾分鐘了', which can cause the skill to activate in unrelated conversations. In a child-focused skill, accidental invocation can lead to unexpected behavior, noisy interruptions, and unintended collection or logging of children's activity data when the user did not mean to use this skill.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The README states that children's brushing records, names, schedules, and reminder state are stored on disk, but it does not warn users about this persistence or the privacy implications. Because the data concerns young children and includes routine information, lack of clear disclosure increases the risk of uninformed consent and exposure of sensitive household behavior if the host account or filesystem is shared or compromised.

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The trigger phrases include very common everyday speech such as '刷牙', '刷牙了沒', and '快去刷牙', which can easily appear in ordinary family conversation. Overly broad triggers can cause accidental invocation, leading to unintended script execution, TTS output, or file updates such as logging and reward tracking.

Missing User Warnings

Low

Confidence: 91% confidence
Finding: The script persistently stores a child-related activity log (name, date, time, duration, session, and derived streak/star data) under the user's home directory without any runtime disclosure, consent prompt, or retention controls. While this is not an exploit primitive by itself, it is a real privacy weakness because sensitive behavioral data can accumulate silently and be exposed to other local processes, backups, or shared accounts.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.