ClawHub
Back to skill

Security audit

01 Tomato Ip Parse

Security checks across malware telemetry and agentic risk

Overview

The skill mostly does the advertised novel analysis, but it sends full manuscript text to a configured model API and falsely returns a passing compliance check without actually checking content.

Review before installing. Use this only with an API_BASE provider you trust for manuscript content, and do not rely on the compliance_check output because the current code always marks content safe instead of performing a real compliance review.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (1)

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill sends raw user-provided novel content and title to an external API endpoint built from environment configuration, with no disclosure, consent check, minimization, or destination validation. This creates a real data-exposure risk because users may submit unpublished or sensitive text, and the code provides no transparency or safeguards before transmitting it off-platform.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.