Back to skill

Security audit

查询技能

Security checks for vulnerabilities and agentic risk

Overview

This is a simple example skill with disclosed dependency and API-key metadata, with no evidence of hidden behavior or malicious actions.

Before installing, read the install commands and only run the sudo package install through your normal trusted admin workflow. Treat this as an example skill and use test or low-privilege API credentials unless you have verified the package source and purpose.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The description field forces a specific language for users reading the skill metadata, and the file does not indicate that the skill is region-specific or that users can choose another language. This is a natural-language policy concern because it may make the skill inaccessible to users who are not expecting Chinese-only descriptions.

Content

No source excerpt is available for this finding.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
88% confidence
Finding

The install instructions include sudo apt-get install -y curl jq, which requires elevated privileges and encourages running package-management commands as root during skill setup. While installing common packages is not inherently malicious, embedding privileged commands in a skill increases risk because users may execute them without scrutiny, and any compromise of the install path or surrounding instructions could have system-wide impact.

Content

Scanner excerpt · SKILL.md (reported line 37)May include surrounding context.

md
- description: "安装 Python 依赖"
        command: "pip install -r requirements.txt"
      - description: "安装系统依赖(Ubuntu/Debian)"
        command: "sudo apt-get install -y curl jq"
        os: linux
      - description: "安装系统依赖(macOS)"
        command: "brew install curl jq"