ClawHub
Back to skill

Security audit

GifHorse

Security checks across malware telemetry and agentic risk

Overview

GifHorse is a coherent media workflow skill, but users should understand it installs an external CLI, indexes local videos, may contact subtitle services, and can optionally send GIFs through iMessage.

Install only if you trust the external GifHorse GitHub repository and are comfortable letting it process the video folders you choose. Use local .srt files if you want to avoid automatic subtitle downloads, keep the transcription database somewhere you control, and verify both the GIF content and recipient before using --send or --send-to.

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 (2)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill explicitly states that transcription may download subtitles automatically and can use Whisper, but it does not clearly warn users that these operations may contact external providers and disclose file names, hashes, subtitle queries, or other metadata derived from the user's local media library. In a privacy-sensitive context, users may reasonably assume a local-only workflow, so the omission can lead to unintended data exposure over the network.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill includes `--send` and `--send-to` features for sending generated GIFs via iMessage, but does not provide a clear warning that this transmits the generated media and recipient phone numbers through messaging infrastructure. Because the content may be derived from a user's private video library, accidental sharing or misuse could expose sensitive media and personal contact information.

VirusTotal

49/49 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.