ppt-icon-generator

Security checks across malware telemetry and agentic risk

Overview

The skill’s command, network, and file-writing behavior is disclosed and fits its UI proof and PR evidence workflow.

Install this only for agents you trust to run repository-local proof commands and publish evidence to GitHub. Prefer dry-runs first, keep dev auth opt-in, and review generated artifacts before publishing them publicly.

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: 96% confidence
Finding: The skill documentation instructs use of shell commands, network access, and writing output files, but the skill declares no permissions. This creates a trust and enforcement gap: an agent or reviewer may assume the skill is low-privilege when it actually performs external requests and filesystem writes, increasing the chance of unsafe execution without appropriate sandboxing or user consent.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal