ClawHub

Model Center

Security checks across malware telemetry and agentic risk

Overview

This appears to be a NVIDIA API integration that sends prompts and text to NVIDIA as expected, but users should treat those inputs as leaving their device.

Install only if you are comfortable sending prompts, image prompts, and embedding text to NVIDIA's API. Do not use it with secrets, private documents, regulated data, or confidential business material unless your NVIDIA account and data-handling terms allow that use.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (6)

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The chat API sends user-provided prompts and conversation content to NVIDIA's external service, but the code path itself provides no disclosure or consent mechanism at the point of transmission. In this skill's context, transmitting prompts to a third-party API is expected functionality, but it still creates a privacy risk because users may unknowingly send sensitive data off-host.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
Image-generation prompts are transmitted to NVIDIA's API without any built-in warning or confirmation to the user at send time. Although remote processing is central to this tool, prompts can contain sensitive or proprietary text, so lack of disclosure increases the chance of unintended data exposure.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
Embedding requests send raw input text to an external API, and there is no user-facing disclosure in this method to indicate that transfer. This is especially sensitive because embeddings are often generated from documents, search queries, or knowledge-base content that may include confidential information.

External Transmission

Medium
Category
Data Exfiltration
Content
}
        
        try:
            response = requests.post(
                endpoint,
                headers=self.headers,
                json=payload,
Confidence
90% confidence
Finding
requests.post( endpoint, headers=self.headers, json=

External Transmission

Medium
Category
Data Exfiltration
Content
}
        
        try:
            response = requests.post(
                endpoint,
                headers=self.headers,
                json=payload,
Confidence
89% confidence
Finding
requests.post( endpoint, headers=self.headers, json=

External Transmission

Medium
Category
Data Exfiltration
Content
}
        
        try:
            response = requests.post(
                endpoint,
                headers=self.headers,
                json=payload,
Confidence
91% confidence
Finding
requests.post( endpoint, headers=self.headers, json=

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal