Back to skill

Security audit

白月光 | Moonlight Map

Security checks for vulnerabilities and agentic risk

Overview

This is a reflective journaling skill for organizing memories about a past person, with clear safety boundaries and no code execution or persistence.

Install only if you want a Chinese-first reflective journaling workflow. Do not treat it as therapy or crisis support, and avoid sharing names, contact details, addresses, workplaces, schools, or other identifying information unless truly necessary.

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 (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The activation examples are broad emotional prompts such as '我忘不了一个人' and '我有个白月光', which can cause the skill to trigger for general heartbreak, rumination, or emotional distress rather than the narrower archival/reflection use case. In an emotionally sensitive skill, over-broad activation increases the chance of inappropriate routing into a structured emotional-analysis workflow when a user may instead need simpler support, clearer scoping, or crisis-aware handling.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

Although the metadata lists both zh-CN and en, the operational instructions, prompts, and required phrasing are overwhelmingly Chinese-centric and do not clearly instruct the agent to ask the user's preferred language before proceeding. This can create consent and comprehension issues, especially in a sensitive emotional context, where users may misunderstand boundaries, safety caveats, or output framing.

Content

No source excerpt is available for this finding.