Security audit

NEXUS Image Analysis

Security checks across malware telemetry and agentic risk

Overview

This appears to be a real paid image-analysis connector, but it needs review because it can send user content to NEXUS and use payment credentials without clear per-use approval rules.

Review before installing. Use the sandbox proof first, configure your agent to ask before sending images or spending payment credentials, and avoid private screenshots, IDs, secrets, faces, or regulated content unless you trust NEXUS and its data-retention practices.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The README states the skill is 'automatically invoked' when a 'matching task is detected' but does not define the trigger conditions, scope, or consent boundaries. In an agent environment, vague auto-invocation language can cause unintended routing of user prompts or images to this external paid service, increasing the chance of privacy leaks and surprise charges.

Missing User Warnings

Medium

Confidence: 97% confidence
Finding: The README describes usage of a remote image-analysis API but does not clearly warn users that prompts, images, and extracted text may be transmitted to an external service. For an image understanding skill, this omission is significant because uploaded images can contain sensitive visual data, embedded documents, faces, locations, or other private information.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.