ClawHub
Back to skill

Security audit

Nano Banana 商品详情页｜PDP 与 Amazon A+

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent AI image-generation helper that uploads user-selected product references to AI Hive and saves generated results locally, with no evidence of hidden publishing, exfiltration, or destructive behavior.

Install only if you are comfortable sending the selected product reference images, prompts, and claims to AI Hive. Use AI_HIVE_API_KEY or --api-key if you do not want the auth command to create ~/.ai-hive/config.json, and avoid uploading confidential or unapproved product assets.

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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill instructs the user to run shell commands, install a Python package, read local reference files, write outputs, use environment variables like SKILL_PATH, and send data to a remote API, but it declares no permissions. This mismatch is dangerous because it hides the true execution and data-handling capabilities of the skill, preventing informed review of filesystem, shell, and network exposure.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The auth command persists the API key to a fixed path on local disk, which creates credential-at-rest risk if the host is shared, backed up insecurely, or later compromised. Although the code attempts to set permissions to 0600, it does not warn the user that the secret will be stored locally or offer safer ephemeral alternatives by default.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.