Back to skill

Security audit

AI Security Guard

Security checks for vulnerabilities and agentic risk

Overview

This is a security-guidance skill that describes command-risk checks and permission patterns; the flagged dangerous commands are examples to block, not instructions to run.

This skill is reasonable to install as reference guidance for command-safety workflows, but treat it as a template rather than a complete enforcement system. Disable sandbox network access unless explicitly needed, avoid bypass or auto modes in untrusted projects, and review any hook or audit-log implementation before enabling it persistently.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Findings (19)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
90% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 32)May include surrounding context.

md
| 模式 | 说明 | 示例 |
|-----|------|------|
| `rm -rf` | 递归删除 | `rm -rf /` |
| `> /dev/sdX` | 磁盘写入 | `echo 1 > /dev/sda` |
| `dd if=` | 裸磁盘操作 | `dd if=/dev/zero of=/dev/sda` |
| `mkfs` | 格式化 | `mkfs.ext4 /dev/sdb` |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
90% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 214)May include surrounding context.

md
| 模式 | 说明 | 示例 |
|-----|------|------|
| `rm -rf` | 递归删除 | `rm -rf /` |
| `> /dev/sdX` | 磁盘写入 | `echo 1 > /dev/sda` |
| `dd if=` | 裸磁盘操作 | `dd if=/dev/zero of=/dev/sda` |
| `mkfs` | 格式化 | `mkfs.ext4 /dev/sdb` |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
90% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 249)May include surrounding context.

md
| 模式 | 说明 | 示例 |
|-----|------|------|
| `rm -rf` | 递归删除 | `rm -rf /` |
| `> /dev/sdX` | 磁盘写入 | `echo 1 > /dev/sda` |
| `dd if=` | 裸磁盘操作 | `dd if=/dev/zero of=/dev/sda` |
| `mkfs` | 格式化 | `mkfs.ext4 /dev/sdb` |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
85% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 32)May include surrounding context.

md
| 模式 | 说明 | 示例 |
|-----|------|------|
| `rm -rf` | 递归删除 | `rm -rf /` |
| `> /dev/sdX` | 磁盘写入 | `echo 1 > /dev/sda` |
| `dd if=` | 裸磁盘操作 | `dd if=/dev/zero of=/dev/sda` |
| `mkfs` | 格式化 | `mkfs.ext4 /dev/sdb` |

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · SKILL.md (reported line 44)May include surrounding context.

md
| 模式 | 说明 | 示例 |
|-----|------|------|
| `curl | sh` | 远程脚本执行 | `curl url \| sh` |
| `chmod 777` | 过度权限 | `chmod 777 /path` |
| `sudo` | 提权操作 | `sudo rm /var/log` |
| `wget` | 远程下载 | `wget -O script.sh url` |

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · SKILL.md (reported line 201)May include surrounding context.

md
| 模式 | 说明 | 示例 |
|-----|------|------|
| `curl | sh` | 远程脚本执行 | `curl url \| sh` |
| `chmod 777` | 过度权限 | `chmod 777 /path` |
| `sudo` | 提权操作 | `sudo rm /var/log` |
| `wget` | 远程下载 | `wget -O script.sh url` |

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · SKILL.md (reported line 44)May include surrounding context.

md
| 模式 | 说明 | 示例 |
|-----|------|------|
| `curl | sh` | 远程脚本执行 | `curl url \| sh` |
| `chmod 777` | 过度权限 | `chmod 777 /path` |
| `sudo` | 提权操作 | `sudo rm /var/log` |
| `wget` | 远程下载 | `wget -O script.sh url` |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 45)May include surrounding context.

md
| 模式 | 说明 | 示例 |
|-----|------|------|
| `curl | sh` | 远程脚本执行 | `curl url \| sh` |
| `chmod 777` | 过度权限 | `chmod 777 /path` |
| `sudo` | 提权操作 | `sudo rm /var/log` |
| `wget` | 远程下载 | `wget -O script.sh url` |
| `pip install` | 包安装 | `pip install unknown` |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 201)May include surrounding context.

md
| 模式 | 说明 | 示例 |
|-----|------|------|
| `curl | sh` | 远程脚本执行 | `curl url \| sh` |
| `chmod 777` | 过度权限 | `chmod 777 /path` |
| `sudo` | 提权操作 | `sudo rm /var/log` |
| `wget` | 远程下载 | `wget -O script.sh url` |
| `pip install` | 包安装 | `pip install unknown` |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
85% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 46)May include surrounding context.

md
|-----|------|------|
| `curl | sh` | 远程脚本执行 | `curl url \| sh` |
| `chmod 777` | 过度权限 | `chmod 777 /path` |
| `sudo` | 提权操作 | `sudo rm /var/log` |
| `wget` | 远程下载 | `wget -O script.sh url` |
| `pip install` | 包安装 | `pip install unknown` |
| `npm i -g` | 全局安装 | `npm i -g package` |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
85% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 214)May include surrounding context.

md
"permissions": {
    "global": {
      "allow": ["echo", "pwd", "ls", "cat"],
      "deny": ["rm -rf /", "> /dev/sda", "dd if="]
    }
  }
}

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
85% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 249)May include surrounding context.

md
import { classifyBashCommand, applyPermissionRules } from './permissions.js'

// 直接分类
const result = classifyBashCommand('rm -rf /tmp/test')
// { decision: 'deny', risk: 'critical', reason: '递归删除' }

// 带上下文的权限判断

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
90% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 249)May include surrounding context.

md
import { classifyBashCommand, applyPermissionRules } from './permissions.js'

// 直接分类
const result = classifyBashCommand('rm -rf /tmp/test')
// { decision: 'deny', risk: 'critical', reason: '递归删除' }

// 带上下文的权限判断

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The activation description is very broad and can trigger on many generic 'security' or 'audit' requests. Over-broad routing increases the chance this skill is invoked in contexts it was not designed for, potentially applying permissive modes, unsafe assumptions, or misleading security guidance to unrelated tasks.

Content

No source excerpt is available for this finding.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 45)May include surrounding context.

md
| 模式 | 说明 | 示例 |
|-----|------|------|
| `curl | sh` | 远程脚本执行 | `curl url \| sh` |
| `chmod 777` | 过度权限 | `chmod 777 /path` |
| `sudo` | 提权操作 | `sudo rm /var/log` |
| `wget` | 远程下载 | `wget -O script.sh url` |
| `pip install` | 包安装 | `pip install unknown` |

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 201)May include surrounding context.

md
| 模式 | 说明 | 示例 |
|-----|------|------|
| `curl | sh` | 远程脚本执行 | `curl url \| sh` |
| `chmod 777` | 过度权限 | `chmod 777 /path` |
| `sudo` | 提权操作 | `sudo rm /var/log` |
| `wget` | 远程下载 | `wget -O script.sh url` |
| `pip install` | 包安装 | `pip install unknown` |

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
70% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 46)May include surrounding context.

md
|-----|------|------|
| `curl | sh` | 远程脚本执行 | `curl url \| sh` |
| `chmod 777` | 过度权限 | `chmod 777 /path` |
| `sudo` | 提权操作 | `sudo rm /var/log` |
| `wget` | 远程下载 | `wget -O script.sh url` |
| `pip install` | 包安装 | `pip install unknown` |
| `npm i -g` | 全局安装 | `npm i -g package` |

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill claims sandbox isolation as a core safety feature, but the default sandbox configuration enables network access. That weakens containment for exactly the high-risk commands this skill is meant to mediate, allowing downloaded payloads, data exfiltration, or outbound callbacks from commands run in the sandbox.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The natural-language description and heading content are entirely in Chinese, but the file does not state that the skill is region-specific or that users may choose another language. This can amount to an implicit language/locale constraint without user opt-in.

Content

No source excerpt is available for this finding.