Security audit

Vulkan Encyclopedia

Security checks across malware telemetry and agentic risk

Overview

This Vulkan helper openly uses official docs and a workspace-local cache, with no evidence of hidden or destructive behavior.

Install this if you want a Vulkan-spec workflow that may create .Vulkan-Encyclopedia files in your workspace. Review the helper scripts before use in sensitive projects, and avoid putting credentials, private URLs, or confidential operational details into the notes.

SkillSpector

By NVIDIA

Vulnerability Patterns

MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 89% confidence
Finding: The skill directs the agent to fetch remote content from docs.vulkan.org and write cached copies and notes into the workspace, yet no permissions are declared. That creates hidden network and file-write side effects that can surprise operators, bypass policy gating, and expand the skill's effective trust boundary even if the target site is legitimate.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 93% confidence
Finding: The skill is presented as a reference workflow for answering Vulkan questions, but it also performs operational actions: remote fetches, local cache creation, and persistence of environment-specific notes. This mismatch can mislead users and policy systems into invoking a skill that modifies the filesystem and consumes network access when they expected read-only analysis.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.