Back to skill

Security audit

AI电商专家|母婴童装电商图片视频

Security checks across malware telemetry and agentic risk

Overview

The skill mostly matches its ecommerce content-generation purpose, but its helper script runs a mutable npm package with broad inherited environment access, which could expose unrelated local secrets.

Install only if you trust IMIVA and the npm package source. Use a dedicated terminal or MCP client profile with only the required MCP_TOKEN and API_URL environment variables, avoid running it from shells that contain cloud or CI secrets, and be aware that local product images or videos you provide may be uploaded to IMIVA for processing and may consume platform credits.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill instructs users to run shell commands, read local files, and consume environment variables for MCP tokens, but it does not declare corresponding permissions or trust boundaries. This creates a capability-transparency gap: users or host systems may expose sensitive local files or credentials without an explicit permission model, increasing the risk of token leakage or unintended local access.

Missing User Warnings

Medium
Confidence
81% confidence
Finding
The script silently executes a remote-fetched package via npx using @latest, which can download and run code from the package registry at invocation time. In an enterprise skill context handling business content and credentials, undisclosed execution of mutable third-party code increases supply-chain and operator-trust risk.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The script prepares an API token and API URL for a network-backed MCP invocation but does not clearly warn the operator that tool arguments and metadata may be transmitted to a remote service. In this skill's ecommerce-content setting, users may provide local file paths, product assets, or sensitive commercial data, making undisclosed transmission more concerning.

Env Variable Harvesting

High
Category
Data Exfiltration
Content
def invoke(method: str, params: dict | None = None) -> dict:
    npx, token = require_runtime()
    env = os.environ.copy()
    env["MCP_TOKEN"] = token
    env["API_URL"] = os.environ.get("IMIVA_API_URL", os.environ.get("API_URL", DEFAULT_API_URL))
    proc = subprocess.Popen(
Confidence
83% confidence
Finding
os.environ.copy()

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.