ClawHub
Back to skill

Security audit

推荐音乐（recommend-music）

Security checks for vulnerabilities and agentic risk

Overview

This is a simple Chinese-language music recommendation skill with no code execution, data access, persistence, or hidden behavior.

Installers should understand this skill is oriented toward Chinese music-recommendation prompts and may answer in Chinese unless the agent adapts language normally. It does not appear to access private data or run commands.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

L003写明“当用户需要音乐推荐、寻找适合特定场景或情绪的音乐、或询问歌曲建议时使用此技能”，其中“询问歌曲建议”范围较宽，且未说明排除条件或边界。该描述缺少负例或更明确的触发约束，容易与一般聊天中较泛化的求建议表达重叠。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

文件整体说明以中文固定描述技能行为，但未声明可根据用户语言偏好切换输出，也未提供用户选择语言的机制。按规则，若技能隐含强制特定语言且没有用户选择或合理限定，属于语言/区域策略问题。

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger section instructs the agent to 'proactively use' the skill for 'any' matching pattern, with examples that are fairly generic. In a multi-skill orchestrator, this can cause overbroad activation on everyday phrases, leading to inappropriate routing, reduced reliability, and possible suppression of safer or more relevant skills when user intent is ambiguous.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

This JSON eval file contains prompts, expected outputs, and assertion conditions entirely in Chinese, including checks that require specific Chinese title text such as '音乐推荐' or '推荐'. That creates a locale/language constraint baked into the skill evaluation without any indication that users may choose another language or that the skill is region-specific.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.