ClawHub
Back to skill

Security audit

数字大脑工厂

Security checks across malware telemetry and agentic risk

Overview

The skill appears to provide the advertised AI analysis service, but it needs Review because it sends potentially sensitive content to external and callback endpoints while also mutating Python import paths outside the reviewed package.

Review before installing. Only submit papers, research ideas, system architecture, logs, or decision context that you are comfortable sending to this provider and to any callback URL you configure. The publisher should remove or justify the home-directory sys.path entries and document privacy, retention, API-key, payment-verification, and callback-safety behavior more clearly.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Tp4

High
Category
MCP Tool Poisoning
Confidence
84% confidence
Finding
The documented purpose says the skill performs analysis and governance services, but the observed behavior includes payment verification, pricing-file reads, and queue/callback placeholder responses that are not disclosed to users. This mismatch undermines user consent and trust, and can cause sensitive system descriptions or research content to be transmitted or processed under assumptions that are materially incomplete.

Context-Inappropriate Capability

Medium
Confidence
87% confidence
Finding
The payment verification trusts caller-supplied fields like 'payStatus' and 'paidAmount' without any cryptographic verification, server-side receipt validation, or provider API confirmation. An attacker can forge a credential object with SUCCESS status and the expected amount to obtain paid functionality without making a real payment.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill explicitly supports remote API submission and callback URLs, but does not clearly warn users that submitted content may be sent to external endpoints and later posted to another server. For a skill handling papers, research ideas, and system audit details, this creates a meaningful confidentiality and data exfiltration risk.

External Transmission

Medium
Category
Data Exfiltration
Content
### 示例1：论文评判
```bash
curl -X POST https://api.v19-cognition.com/analyze -H "Content-Type: application/json" -d "{\"task_type\":\"paper_review\",\"content\":\"机器学习在NLP中的应用研究\",\"depth\":4}"
```

### 示例2：系统审计
Confidence
90% confidence
Finding
curl -X POST https://api.v19-cognition.com/analyze -H "Content-Type: application/json" -d "{\"task_type\":\"paper_review\",\"content\":\"机器学习在NLP中的应用研究\",\"depth\":4}" ``` ### 示例2：系统审计 ```bash curl -

External Transmission

Medium
Category
Data Exfiltration
Content
```bash
# 验证协议
curl -s http://127.0.0.1:8700/governance/health
# 对偶审计
curl -X POST http://127.0.0.1:8700/governance/audit -H "Content-Type: application/json" -d '{"decision_id":"d001","context":"测试","chosen_action":"验证","evidence":"ok"}'
# 因果归因
Confidence
88% confidence
Finding
curl -s http://127.0.0.1:8700/governance/health # 对偶审计 curl -X POST http://127.0.0.1:8700/governance/audit -H "Content-Type: application/json" -d '{"decision_id":"d001","context":"测试","chosen_action":

External Transmission

Medium
Category
Data Exfiltration
Content
### 示例1：论文评判
```bash
curl -X POST https://api.v19-cognition.com/analyze -H "Content-Type: application/json" -d "{\"task_type\":\"paper_review\",\"content\":\"机器学习在NLP中的应用研究\",\"depth\":4}"
```

### 示例2：系统审计
Confidence
90% confidence
Finding
https://api.v19-cognition.com/

External Transmission

Medium
Category
Data Exfiltration
Content
### 示例2：系统审计
```bash
curl -X POST https://api.v19-cognition.com/analyze -H "Content-Type: application/json" -d "{\"task_type\":\"system_audit\",\"content\":\"我的AI系统包含支付、认知、执行模块，请审计其完整性\",\"callback_url\":\"https://my-server.com/result\"}"
```

## 错误处理
Confidence
95% confidence
Finding
https://api.v19-cognition.com/

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.