ClawHub
Back to skill

Security audit

RouterBase Media Generation

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only RouterBase media integration skill whose API-key, network, async polling, and media storage guidance matches its stated purpose.

Before installing, confirm you are comfortable sending prompts, input media URLs, and generated media requests to RouterBase. Store generated files and job metadata only where your application has appropriate user consent, access controls, retention limits, and deletion handling, and avoid logging API keys or private media URLs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
This markdown file documents storing async job metadata, including the full request payload and result URLs, which may contain user prompts or user-generated media references. Under the markdown-specific warning criterion, the description should warn that retained data may affect user privacy or require careful handling, but no such warning is present here.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The file instructs applications to download and store generated media on their side, which can materially affect user data handling and privacy. For markdown files, behavior that impacts user data should be accompanied by a warning, but this section gives operational guidance without any disclosure about storage, retention, or user consent implications.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.