ClawHub
Back to skill

Security audit

LTX-2.3 Video API

Security checks across malware telemetry and agentic risk

Overview

This video-generation skill is mostly coherent, but it exposes a live-looking API key and recommends sending private media to a public file host without adequate warning.

Review before installing. Do not use the embedded API key; it should be treated as exposed and rotated by its owner. Use your own LTX key through an environment variable or secret manager, and avoid uploading sensitive portraits, voice recordings, or private videos to public file hosts unless you understand their retention and access policies.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The skill directs users to upload audio and images to uguu.se, an unrelated third-party file host, as part of the workflow. This expands data exposure beyond the stated LTX API purpose and can leak sensitive media to an uncontrolled external service, especially because the content may include portraits or voice recordings.

Missing User Warnings

High
Confidence
100% confidence
Finding
The skill exposes what appears to be a real API key directly in documentation and tells users where it is stored locally. This is an active secret disclosure that can enable unauthorized API use, account abuse, billing fraud, and lateral compromise if the same environment or storage practices are reused elsewhere.

Missing User Warnings

High
Confidence
97% confidence
Finding
The skill recommends uploading potentially sensitive user media to a third-party file host without any privacy, retention, or access-control warning. In this skill's context, the data may include faces and voice recordings, making the omission especially dangerous because users may unknowingly disclose biometric or personal content.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.