skill-tester-cn

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 6:56 AM UTC
• Version: 1.0.1

Overview

This is a coherent Chinese-language skill-testing helper that reads target skill definitions, runs or simulates tests, and writes a report, with no evidence of hidden persistence, exfiltration, or destructive behavior.

Install this only if you want a Chinese-language helper that can inspect local skill definitions and create Markdown test reports. When testing powerful skills, prefer simulated tests or review prompts first, because actual execution may trigger the target skill's normal side effects.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (9)

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The documented trigger phrases are broad, natural-language expressions such as '评估技能' and '检查技能是否工作', which can plausibly appear in ordinary conversation and unintentionally activate the skill. Because this skill appears capable of locating other skills, generating tests, and producing reports, accidental invocation could cause unintended file reads and write-side effects.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The README states that the skill will execute tests and generate a report, but it does not clearly warn users that this may read target skill files and create output artifacts on disk. This lack of disclosure increases the risk of users triggering operations with side effects they did not expect, especially in environments containing sensitive or important local skill data.

Skill Enumeration

Medium

• Category: Agent Snooping

Content

用户: 测试 docx 技能

Claude 将：
1. 定位技能文件 ~/.claude/skills/docx/SKILL.md
2. 解析技能定义，识别核心能力
3. 生成测试用例（创建、编辑、提取等）
4. 执行测试并记录结果

• Confidence: 80% confidence
• Finding: Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Skill Enumeration

Medium

• Category: Agent Snooping

Content

用户: 测试 docx 技能

Claude 将：
1. 定位技能文件 ~/.claude/skills/docx/SKILL.md
2. 解析技能定义，识别核心能力
3. 生成测试用例（创建、编辑、提取等）
4. 执行测试并记录结果

• Confidence: 80% confidence
• Finding: Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Natural-Language Policy Violations

Medium

• Confidence: 95% confidence
• Finding: The skill name and description explicitly position the skill as Chinese-language ("cn" and Chinese-only instructions), with no indication that users may choose another language or that the locale restriction is required for compliance or regional use. This is a natural-language locale policy issue because it imposes a specific language without opt-in.

Skill Enumeration

Medium

• Category: Agent Snooping

Content

```
用户说: "测试 PDF 技能"
→ 搜索: ~/.claude/skills/pdf/SKILL.md 或 pdf/SKILL.md
```

常见技能位置：

• Confidence: 80% confidence
• Finding: Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Natural-Language Policy Violations

Medium

• Confidence: 95% confidence
• Finding: The template's headings and fixed text are entirely in Chinese, including the generated-by footer, which implies reports are produced in a specific language by default. This can violate language/locale policy when no user opt-in, selection mechanism, or documented region-specific justification is provided.

Natural-Language Policy Violations

Low

• Confidence: 90% confidence
• Finding: README 从标题和正文开始即明确以中文描述技能，但没有说明这是可选语言、提供多语言支持，或解释该技能仅面向特定中文环境。若组织要求避免未经用户同意强制特定语言，这属于自然语言层面的语言策略风险。

Missing User Warnings

Low

• Confidence: 92% confidence
• Finding: The skill instructs writing a generated report into the current working directory without requiring an explicit user confirmation or warning that local files will be created. This can cause unintended workspace modification, clutter, or accidental overwriting if naming collisions occur, especially when run in sensitive project directories.

Static analysis

No suspicious patterns detected.