Back to skill

Security audit

Daily Sleep

Security checks for vulnerabilities and agentic risk

Overview

This sleep-coaching skill is purpose-aligned and its recurring notification behavior is visible, though users should understand that enabling pushes creates scheduled messages through third-party channels.

Install only if you want a sleep assistant that can create recurring morning and evening reminders. Before enabling pushes, choose the intended channel and user ID carefully, and use the documented off command if you no longer want scheduled messages.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
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 (12)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

This skill presents itself as a content-oriented sleep assistant but exposes operational commands for enabling, disabling, and checking push jobs across external messaging platforms. That undeclared infrastructure capability broadens the skill’s real function and can create privacy and consent issues if users are enrolled in reminders or their messaging destinations are managed without sufficiently clear notice.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

This skill presents itself as a content-oriented sleep assistant but exposes operational commands for enabling, disabling, and checking push jobs across external messaging platforms. That undeclared infrastructure capability broadens the skill’s real function and can create privacy and consent issues if users are enrolled in reminders or their messaging destinations are managed without sufficiently clear notice.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The README promotes scheduled morning and evening push notifications and shows commands to enable them, but it does not warn users that notifications are recurring, may be delivered over third-party channels, and may reveal sleep-related habits or preferences. Because this is a wellness/sleep skill with daily schedules and multi-channel delivery, missing disclosure raises consent, privacy, and nuisance risks if users enable pushes without understanding persistence and channel effects.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list includes very broad, common phrases such as 'sleep', 'insomnia', 'can't sleep', and generic Chinese sleep-related terms, which increases the chance of unintended invocation during ordinary conversation. In a health-adjacent coaching skill, accidental activation can cause unsolicited advice, unnecessary collection or routing of user context, and user confusion across supported messaging channels.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger phrases are broad enough to match routine conversation about sleep, which can cause unintended activation of the skill. In a skill that can manage push notifications, accidental invocation is more sensitive because it may lead users into enabling reminder workflows they did not explicitly seek.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The description advertises a recurring, broadly scoped behavior ('Daily evening push for better rest') without clearly stating the user-controlled trigger, opt-in conditions, or delivery constraints. In an agent ecosystem, vague proactive activation language can enable unwanted notifications or repeated engagement patterns that affect user autonomy and create abuse potential even if the feature is framed as wellness-related.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The manifest explicitly promotes proactive daily pushes but does not warn that the skill may send recurring notifications or explain how user impact is controlled. Even for a sleep-focused skill, this can mislead reviewers or users about engagement behavior and normalize attention-grabbing patterns that become intrusive if enabled by default or insufficiently consented.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The script hard-codes Chinese weekday names and a fully Chinese prompt, requiring the generated sleep guidance to be in Chinese. This is a natural-language locale policy issue because the file provides no user opt-in, locale selection, or explanation that the skill is intended only for Chinese-speaking users.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The script defines weekday names in Chinese and emits a fully Chinese prompt, which forces a specific language/locale on all users. This is a natural-language policy concern because there is no opt-in, fallback, or documented region-specific justification in the file.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The manifest describes a sleep skill centered on nighttime wind-down routines, breathing exercises, bedtime stories, sleep hygiene tips, and a daily evening push. This script instead emits a morning wake-up prompt ('早安') for a '晨间唤醒引导' with activation steps, which is a different user-facing behavior from the declared evening sleep support scope.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The script prints user-facing status and confirmation messages in Chinese at L25, L32, and L37. This imposes a specific language on all users without opt-in or any indication that the skill is intended only for Chinese-speaking or region-specific use, which fits the language/locale policy violation category.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
72% confidence
Finding

The skill presents its title and main usage guidance in Chinese while also including English keywords and trigger phrases, but it does not state how language selection is handled for users. This may create a locale-policy concern if the skill defaults to one language experience without explicit user choice.

Content

No source excerpt is available for this finding.