ClawHub
Back to skill

Security audit

fame graphic

Security checks across malware telemetry and agentic risk

Overview

This skill coherently generates illustrations through the OpenAI Images API and saves local outputs, with privacy and local gallery safety caveats users should understand.

Install only if you are comfortable sending illustration prompts and related parameters to OpenAI or a configured OpenAI-compatible endpoint. Use a scoped API key, verify OPENAI_BASE_URL or OPENAI_API_BASE before running, avoid putting secrets or confidential material in prompts, and be cautious opening generated index.html galleries from untrusted prompt content.

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 requires environment access for `OPENAI_API_KEY` and network access to call the OpenAI Images API, but it does not declare those permissions. This creates a transparency and governance gap: users or orchestration systems may invoke a networked skill with secret access without being clearly informed or able to apply permission controls.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill description explains how to generate illustrations but does not warn that user-provided prompts/content are transmitted to the OpenAI Images API and that generated images and prompt metadata are written locally (including `prompts.json` and `index.html`). This can expose sensitive user content to an external service and leave local artifacts containing prompts or generated material without the user's explicit awareness.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.