Back to skill

Security audit

Local Image Upscaler

Security checks across malware telemetry and agentic risk

Overview

This is a coherent local image upscaling and compression skill that discloses its downloads, local file access, cache use, and safety limits.

Install this if you are comfortable with a local image tool downloading and running a pinned third-party Upscayl binary and model files. Use trusted mirrors only, keep offline mode for prefilled caches, and review output paths before batch jobs or using --overwrite.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill instructs the agent to use shell commands, read and write local files, access environment variables, and optionally use the network for downloads and mirrors, yet no permissions are explicitly declared. This creates a trust and review gap: users and platform controls may not understand that the skill can access local paths and fetch remote artifacts, increasing the chance of unintended file access or network exposure.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.