Xiaozhi Time Focus Coach

• Outcome: Pass
• Latest audit: Jun 6, 2026 at 6:23 AM UTC
• Version:
  1.0.1

Overview

This skill is a study focus coach that openly tracks study time and distractions for coaching, with no executable code or hidden install behavior found.

Install this if you want a coach that remembers study sessions and distraction patterns over time. Before using long-term tracking, confirm what will be stored, where it is stored, how to view or delete it, and whether summaries may be shared with related learning-plan or reminder skills, especially if the user is a minor.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (4)

Vague Triggers

Medium

• Confidence: 96% confidence
• Finding: The skill declares that it 'must' activate for a broad set of common student statements such as lacking time, procrastination, distraction, or wanting help tracking study time. These phrases overlap heavily with ordinary conversation, so the agent may invoke the skill when the user did not intend persistent tracking or workflow changes, creating consent, privacy, and routing risks.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The instruction that any scenario involving study-time recording, focus training, or task optimization should invoke this skill is overly broad and lacks clear boundaries. In practice, this can cause over-collection of behavioral data or unnecessary skill routing in contexts where the user only asked a general question, increasing the chance of unintended persistence or cross-skill sharing.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The state machine explicitly says the AI will update a persistent 'focus profile/DNA' and archive completion, abandonment, duration, and distraction records, but this file does not require a user-facing notice or consent at the point of collection. Because the skill targets students and behavioral data is longitudinal, silent retention can create privacy harm, profiling risk, and unexpected monitoring of sensitive habits over time.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The workflow instructs recording detailed distraction-event metadata such as time, task type, interruption reason, and interruption duration, yet it does not require warning the user that these behavioral details will be logged and persisted. This creates a transparency and privacy issue because the system is building a detailed behavioral dataset that could reveal routines, device use, and attention patterns without informed consent.

VirusTotal

64/64 vendors flagged this skill as clean.