Security audit

iPhone 随手拍(中文)

Security checks for vulnerabilities and agentic risk

Overview

This is a prompt-only skill for generating Chinese iPhone-style image prompts, with disclosed language and style behavior and no executable code or hidden system access.

Install this skill if you want image prompts to default to Simplified Chinese and an iPhone candid-photo style. Be explicit when you want English output or a non-phone, studio, DSLR, or cinematic look so the skill does not steer the prompt toward its default style.

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

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (8)

Natural-Language Policy Violations

Medium

Confidence: 92% confidence
Finding: The notice states that the modified version changes the output language so prompts are output in Chinese, indicating a fixed language behavior. Under the policy, forcing a specific language without offering user choice or opt-in is a natural-language policy violation.

Natural-Language Policy Violations

Medium

Confidence: 97% confidence
Finding: Declaring Chinese-only output by default imposes a behavior constraint before the user has opted into that language preference. In a multi-skill or multilingual agent environment, this can silently override user expectations, reduce reliability, and interfere with downstream tools that expect another language or locale-specific formatting.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The README says the skill can be triggered by broad natural-language cues such as requests for candid, phone-shot, 自拍, plandid, or simply wanting Chinese image prompts. Overbroad activation criteria can cause the skill to override user intent in unrelated image-generation tasks, leading to prompt hijacking at the skill-selection layer and unexpected behavior without explicit consent.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: L03 将“随手拍、手机实拍质感、抓拍、plandid、生活化自拍、路人视角照片、想要中文图像提示词或 JSON 提示词”统称为应使用该 skill 的触发条件，其中多项属于宽泛风格描述或常见生成需求，而不是足够具体的技能调用边界。虽然给出了一个负向排除条件，但仍缺少更明确的触发范围或反例，容易在普通图片生成请求中被过度匹配。

Natural-Language Policy Violations

Medium

Confidence: 97% confidence
Finding: L03 规定即使用户没有明确说“中文”，提示词也默认输出简体中文，这会在未获得用户明确选择的情况下强制施加语言偏好。该文件虽在 L24 允许用户明确要求英文时切换，但默认策略仍是预设语言而非用户自选。

Natural-Language Policy Violations

Medium

Confidence: 93% confidence
Finding: L22-L25 将输出语言规则设为所有字段一律使用简体中文，仅在用户明确要求英文时例外。这属于默认强制特定语言的行为，而不是提供等价的语言选择或基于用户偏好的自适应。

Natural-Language Policy Violations

Medium

Confidence: 97% confidence
Finding: The skill hard-codes Chinese output in its display text and default prompt, and the metadata states Chinese should be the default even without explicit user opt-in. This can override user language expectations, reduce transparency, and create downstream safety/usability issues if users cannot understand the generated prompt or if other components assume output will follow the user’s language.

Natural-Language Policy Violations

Medium

Confidence: 95% confidence
Finding: The title and instructions require Chinese output, including '中文输出' and '除专有名词外全部用简体中文书写', which imposes a specific language on users. This is a natural-language policy concern because the file does not provide an opt-in mechanism or explain a region-specific need for the language constraint.

Static analysis

No suspicious patterns detected.