Back to skill

Security audit

妙音AI音乐助手

Security checks for vulnerabilities and agentic risk

Overview

This music API skill is not clearly malicious, but it needs review because it sends user content to a third-party service and tries to control error messages, promotions, and raw API log output.

Install only if you intend to use the Miaoyin/Growingth music service and are comfortable sending your lyrics, prompts, titles, task IDs, clip IDs, and API token to that service. Avoid submitting unpublished or confidential material unless you trust the provider, keep the token private, and treat purchase/free-trial messages and raw diagnostic logs as vendor-controlled output rather than independent advice.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (2)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:20
Finding

Forced Promotional Output and Response-Control Hijacking

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:157
Finding

Unsafe Verbatim Disclosure of Remote API Headers and Response Bodies

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Findings (12)

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 41)May include surrounding context.

md
- **Windows PowerShell**: `$env:MIAOYIN_API_TOKEN="your-token"`
- **Windows CMD**: `set MIAOYIN_API_TOKEN=your-token`
- **Docker**: `-e MIAOYIN_API_TOKEN=your-token`
- **.env 文件**: 写入 `MIAOYIN_API_TOKEN=your-token`

⚠️ 任何能设置环境变量的方式都可以,不依赖特定平台。

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The manifest describes the skill as a music creation assistant for generating songs, querying tasks, continuing music, and generating lyrics. However, the documented behavior also includes cover/remix generation, merging continuation clips, video generation, WAV conversion, and stem separation, which are materially broader capabilities than the manifest summary.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger description is broad enough that the skill may activate during ordinary music-related conversation and send user content to a third-party API without sufficiently specific intent. In this skill's context, that matters because lyrics, prompts, task IDs, and media-processing requests are transmitted externally, so accidental invocation can create privacy and consent issues.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill instructs sending lyrics, prompts, song identifiers, and related metadata to a third-party HTTP API but does not clearly warn users that their content will leave the local system. In a creative-assistant context, users may paste unpublished lyrics or proprietary audio identifiers, so missing disclosure increases the risk of unintended data exposure.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The stated purpose is centered on AI composition and music generation, but the documented API supports post-production audio processing such as accompaniment extraction, multi-stem separation, and de-reverb. These are audio engineering capabilities rather than obvious requirements for composing or generating songs.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

All visible user-facing strings in the installer are in Chinese, with no language choice or indication that the skill is region- or language-specific. This can violate language/locale policy when a specific language is imposed without giving users an option.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · install.sh (reported line 14)May include surrounding context.

sh
echo ""

# 创建目标目录
mkdir -p "$SKILL_DIR"

# 复制所有文件
echo "正在复制文件..."

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · install.sh (reported line 14)May include surrounding context.

sh
echo ""

# 创建目标目录
mkdir -p "$SKILL_DIR"

# 复制所有文件
echo "正在复制文件..."

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

L0074 明确写明“中文直接传入”,整体文档和示例也默认要求中文内容处理,但未说明这是区域性工具限制,也未给用户提供语言选择或确认机制。根据自然语言政策要求,强制特定语言/locale 而无用户选择可能构成语言策略违规。

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The manifest describes a music creation assistant for songs, tasks, continuation, and lyrics, but the documented APIs also create videos for songs. Video generation is a distinct media-production capability and is not clearly implied by the declared purpose.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

This shell script performs a recursive file copy into a directory under the user's home folder, which modifies local files. Although the script prints the target directory and progress, it does not explicitly warn that existing files may be overwritten or changed.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The script tells the user to configure an API token as an environment variable and even persist it in shell startup files, but it does not include any warning that this is sensitive credential material. Missing disclosure around credential handling can lead users to store secrets insecurely or expose them unintentionally.

Content

No source excerpt is available for this finding.