Back to skill

Security audit

Lucky Wallpaper Generator

Security checks for vulnerabilities and agentic risk

Overview

The skill is not clearly malicious, but it overstates its image-generation capability and describes external API use and automatic publishing without enough user-control detail.

Review this skill before installing. It appears to be a prompt/config generator rather than a complete image generator, and any use of external image APIs or Xiaohongshu publishing should require explicit user confirmation, account selection, and content review. Avoid very large generation counts until the script enforces limits.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/generate.js:88
Finding

Unbounded Generation Count Allows Local Resource Exhaustion

Content
View full analysis
Remediation
View remediation
MAX_COUNT) { throw new RangeError(`count must be an integer between 1 and ${MAX_COUNT}`); } } ``` Invoke this validation before creating the output directory or entering the generation loop. In batch mode, additionally verify that `types.length * countPerType` does not exceed a fixed total-generation limit. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • 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
Findings (13)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

描述声称该技能能够“自动生成”壁纸、头像和图卡,听起来是直接产出图片内容;而实际代码只是本地创建输出目录,随机选取预设英文prompt,并将类型、尺寸、prompt、时间戳等写入JSON文件,状态还标记为“ready_for_generation”,且明确写有“请使用通义万相或即梦API生成图片”。这表明代码的真实功能是生成图片生成任务配置,而不是生成最终图片。四大系列和多尺寸支持与描述部分一致,但核心能力存在明显夸大,因此应判定为描述与实际行为不符。

Content

No source excerpt is available for this finding.

Direct Prompt Extraction

High
Category
System Prompt Leakage
Confidence
85% confidence
Finding

Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Content

Scanner excerpt · scripts/generate.js (reported line 67)May include surrounding context.

js
throw new Error(`未知的类型: ${type}`);
  }
  const prompts = template.prompts;
  return prompts[Math.floor(Math.random() * prompts.length)];
}

// 生成壁纸

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The string "请使用通义万相或即梦API生成图片" directs use of specific Chinese services and is written only in Chinese, which can impose a language/locale constraint without user opt-in. Under the policy rule, this is a natural-language locale preference that is not presented as optional or justified as region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The package description states the skill generates '小红书风格' wallpapers and related assets in Chinese, which implies a fixed language/locale orientation. For all file types, locale-specific behavior should either offer user opt-in or clearly document a justified regional constraint; this manifest does neither.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The top-level comment says '自动生成小红书风格的好运壁纸', which implies the script itself creates wallpaper outputs. However, the implementation writes JSON records marked 'ready_for_generation' and includes an 'apiHint' telling the user to use an external image API, contradicting the stated behavior in the comment.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest description presents the skill as directly generating '好运壁纸、招财头像、情绪图卡'. In this implementation, the main generation function only assembles metadata and prompts, then writes them as .json files with status 'ready_for_generation' and an API hint instructing a separate image API to do the real generation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill states that it integrates with third-party image-generation APIs but does not disclose that user prompts or related content may be transmitted to external services. This creates a privacy and data-handling risk, especially if users include personal, commercial, or sensitive text in generation requests.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill advertises automatic publishing and later describes saving outputs, but it does not clearly warn users that content may be written to disk or posted to an external platform. This can cause unintended data persistence or unauthorized publication if a user assumes generation is local and non-persistent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

该 markdown 文件适用 SQP-1。使用方法中的触发示例如“生成一张招财壁纸”“生成治愈系头像”属于自然日常表达,且文档未说明这些是否是精确触发词、是否需要特定上下文,或哪些相似表达不应触发技能,因此存在误调用风险。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The string literal at L16 instructs use of image-generation APIs in Chinese only, which can constitute a language/locale policy issue when no user opt-in or alternative language is provided. Because this is a manifest-style JSON file and no regional justification is present elsewhere in the file, the guidance appears to impose a specific language context by default.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

The natural-language value at L16 instructs use of image-generation APIs entirely in Chinese, which imposes a specific language/locale in the skill metadata. The file does not offer an alternative language or explain that the skill is intentionally limited to a Chinese-language environment.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

SQP-3 适用于所有文件类型。该文件的标题、说明、示例与操作描述均默认采用中文,且未说明是否支持其他语言或允许用户选择语言;若组织要求避免未获用户同意的语言限定,这构成自然语言层面的潜在策略问题。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

This JSON contains multiple user-facing names, descriptions, and text strings in Chinese, including display labels for themes and output text, while providing no indication that language is selectable or that the skill is intentionally limited to a Chinese-speaking audience. Under the policy for natural-language constraints, forcing a specific language without user opt-in can be a locale-policy issue.

Content

No source excerpt is available for this finding.