Security audit

lux3d

Security checks across malware telemetry and agentic risk

Overview

This is a coherent Lux3D API client, but it needs Review because it can send user content and the Lux3D API key to an arbitrary override endpoint.

Install only if you are comfortable sending selected prompts, images, and model URLs to Lux3D. Avoid using sensitive or proprietary assets unless you have checked Lux3D’s data terms, and do not set LUX3D_BASE_URL or --base-url unless you fully trust that endpoint because it will receive your API key and submitted content.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection

Findings (7)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 95% confidence
Finding: The skill documentation describes capabilities that use environment variables, write output files, and send data over the network, but no declared permissions are present. This creates a transparency and governance gap: users and the hosting platform may not realize the skill can exfiltrate prompts, images, and URLs to an external service and persist downloaded artifacts locally.

Description-Behavior Mismatch

Medium

Confidence: 97% confidence
Finding: The client permits an arbitrary LUX3D_BASE_URL or --base-url override and then uses that value to construct authenticated API requests. Because local images, prompts, mesh URLs, and the Authorization header are all sent to the configured endpoint, this expands the trust boundary from the documented Lux3D service to any host, enabling credential and data exfiltration if the override is influenced by an attacker or misconfiguration.

Context-Inappropriate Capability

High

Confidence: 99% confidence
Finding: When LUX3D_BASE_URL is set, the code returns it directly and later attaches the API key in the Authorization header for all requests. This means any attacker-controlled or accidentally unsafe endpoint can receive both sensitive user inputs and the bearer credential, creating a direct exfiltration channel outside the vendor's stated infrastructure.

Vague Triggers

Medium

Confidence: 83% confidence
Finding: The trigger description is broad enough to match many ordinary 3D-generation requests, increasing the chance the skill activates without a clear user intent to use Lux3D specifically. That can cause unintended transmission of user-provided content to the external API and unexpected file generation or downloads.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The skill explains workflows and required inputs but does not clearly warn that images, prompts, and model URLs are transmitted to a third-party API. Users may provide sensitive or proprietary assets without informed consent, creating privacy, confidentiality, and compliance risks.

Missing User Warnings

Medium

Confidence: 78% confidence
Finding: The image-to-3D flow base64-encodes local images and submits them to a remote API, but the upload path itself provides no explicit user-facing notice or consent checkpoint. In a skill context handling local files, this can lead to unintended disclosure of private or sensitive images if users do not realize the content leaves the local environment.

Missing User Warnings

Medium

Confidence: 79% confidence
Finding: The text-to-3D workflow sends user prompts and optional reference images to a remote API without an explicit warning at the submission point. Prompts and images may contain sensitive business, personal, or proprietary information, so silent transmission to a third party creates a real privacy and data-governance risk.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.