ClawHub
Back to skill

Security audit

反无限循环守护

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small Chinese-language loop-guard helper with no evidence of hidden access, persistence, network use, or destructive behavior.

Install this only if you want a Chinese-language helper for adding simple loop limits and repeated-action checks. Treat its performance and resource-protection claims as lightweight guidance rather than a complete enforcement system.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (6)

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding
声明描述的是一个用于防止代理陷入无限循环的守护型技能，应当具备检测循环、判断重复、跟踪执行状态或触发终止条件等核心行为。但给出的代码只是一个简单的命令行脚本骨架：创建参数解析器、接受一个verbose标志、输出固定版本字符串并退出。代码没有读取任务状态、没有比较历史操作、没有计数器或超时、没有终止机制、没有资源监控，也没有任何与触发词相关的实际实现。因此其实际行为与声明用途存在明显且实质性的偏差。

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding
从提供的代码看，`setup_logging` 只是配置日志，`get_skill_root` 返回目录路径，`validate_environment` 仅报告 Python 版本并固定返回有效，`log_metrics` 只是打印指标文本。这些都是通用辅助功能，不足以支持“防止代理陷入重复执行循环”的主要能力。虽然 `log_metrics` 与“监控/进度”有弱相关性，但并没有实现任何进度状态管理、循环次数统计、重复模式检测或终止机制。因此该代码块的实际行为与声明用途存在明显且实质性的偏差。

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding
声明描述的是一个用于防止代理陷入无限循环的功能性技能，但实际代码只是 tests/test_basic.py 中的两个 assert True 测试，占位性质明显。该代码既不包含业务实现，也未展现任何与循环监控、重复操作识别、终止判定或进度追踪相关的行为。因此，代码实际行为与声明用途存在明显且实质性的不一致。

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
The skill name, description, and user-facing content are entirely in Chinese, and the file does not indicate that language selection is optional or that the skill is intentionally limited to a Chinese-speaking context. This creates a natural-language policy concern because it effectively enforces a locale without user opt-in.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrase '防止无限循环' is broad enough that normal user requests about debugging, optimization, or general execution control could activate this skill unintentionally. Over-broad activation can cause inappropriate intervention in unrelated tasks, leading to denial of intended behavior, workflow disruption, or masking of the correct skill.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger phrase '重复操作检测' is ambiguous and could match many benign contexts such as UI automation, data deduplication, or audit tasks. In an agent system, ambiguous activation increases the chance of accidental skill invocation, which can alter execution flow and create reliability or availability issues.

Static analysis

No suspicious patterns detected.