Use when user wants to review material forms for data sharing catalogs, field completeness, platform consistency, and issue-list output. Triggers include「材料审核」「共享清单审核」「检查文档审查」「平台对接核对」「编目一致性检查」.

Security checks across malware telemetry and agentic risk

Overview

This skill appears safe to install: its local file output is disclosed, user-directed, and fits the stated reporting/audit purpose.

Before installing, confirm where the output directory points, avoid paths that could overwrite important files, and keep generated reports in a private location if the submitted material contains sensitive data.

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: 91% confidence
Finding: The skill advertises no explicit permissions, but its documented workflow writes audit artifacts to disk via `--output-dir` and produces multiple files. This creates a capability/permission mismatch: a host may approve or sandbox the skill assuming it is read-only, while the skill can persist data derived from submitted materials, increasing the risk of unauthorized file creation, data leakage, or overwrite of local content if path handling is weak elsewhere.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal