Security audit

Qa Security

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed paid QA/security-audit workflow that creates and verifies a payment order before the agent delivers the requested QA or security advice.

Before installing, expect a Chinese-language paid workflow that contacts https://api.ideaidea.com.cn, stores local order metadata under OpenClaw's orders directory, and sends a payment credential from that order file for verification only after confirmation. Do not use it unless you are comfortable with that payment provider and workflow.

SkillSpector

By NVIDIA

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (8)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 70% confidence
Finding: Without declared permissions the skill's intent is opaque and cannot be validated.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 98% confidence
Finding: 描述将技能定位为执行 QA/测试/安全审计工作的服务技能，但代码并没有实现任何测试计划、代码审查、漏洞分析、风险评估或质量闸口相关逻辑。其核心行为是支付流程支撑：调用固定域名的支付 API、保存订单 JSON 文件、再读取本地订单中的 payCredential 去校验支付结果。虽然描述中提到“付费”场景，且 metadata 里有 payment.process，但声明的技能用途仍然是测试与安全审计本身，而不是一个纯支付下单/验单工具。因此代码的主要目的与描述存在实质性偏差，应判定为不匹配。

Context-Inappropriate Capability

Medium

Confidence: 93% confidence
Finding: The manifest describes the skill as a paid testing and security audit service, with payment presented as a prerequisite workflow step rather than the core user-facing purpose. Declaring `payment.process` as a capability overstates what the skill is fundamentally for, because the actual service scope is QA and security assessment, not payment operations.

Description-Behavior Mismatch

Medium

Confidence: 95% confidence
Finding: The manifest describes a paid testing and security audit skill for activities like test planning, code review, vulnerability investigation, risk assessment, and release gating. This module instead implements persistent order storage under a user home directory, including loading and saving order JSON files, which is a business/payment workflow rather than an obvious implementation detail of QA or security auditing.

Context-Inappropriate Capability

Medium

Confidence: 93% confidence
Finding: The code provides a concrete capability to manage local order records keyed by an order number and indicator, including reading existing orders and writing new ones. For a skill whose stated purpose is one-off testing and security audit delivery, this order-management/storage capability is not directly justified by the described operational scope.

Natural-Language Policy Violations

Medium

Confidence: 97% confidence
Finding: The markdown explicitly states '请使用中文和用户交互', which mandates a specific language for all user interaction. This is a natural-language locale policy issue because it does not offer the user a choice or describe a justified region-specific constraint.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The save_order function creates directories and writes JSON order data to a fixed path under the user's home directory, which affects local user data. In this file, there is no confirmation prompt, logging/print statement, or explicit user-facing warning about this filesystem write behavior.

Natural-Language Policy Violations

Low

Confidence: 81% confidence
Finding: Comments, docstrings, and error messages such as the Chinese-language descriptions and runtime text indicate a single fixed language experience. Under the policy, forcing a specific language without offering user choice or documenting a justified locale constraint can be a natural-language policy violation.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.