Back to skill

Security audit

deep-night-treehole-1.0.0

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent Chinese emotional-support skill with no code execution, data access, persistence, or hidden authority.

Install this if you want a Chinese-language emotional companion style. Be aware it may activate on common distress-related phrases and it is not a substitute for professional mental health support in serious situations.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

The README defines the skill entirely in Chinese and does not indicate any language negotiation or user-language fallback behavior. In multilingual environments, this can cause the assistant to respond in an unexpected language during emotionally sensitive interactions, reducing clarity and trust and potentially worsening user experience at critical moments.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill is configured to auto-activate on broad emotional keywords such as '累了', '焦虑', and '想聊聊', which can easily appear in ordinary conversation without the user intending to invoke an emotional-support persona. This can cause unintended behavior changes, including inappropriate emotional framing, missed routing to more suitable skills, or over-engagement in sensitive contexts where a more explicit opt-in would be safer.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly instructs the agent to use warm, conversational Chinese without offering any user language choice. This can create accessibility and consent issues by forcing interactions into a language the user may not understand, which is especially risky in an emotional-support context where misunderstandings can worsen distress or prevent users from receiving appropriate help.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.