Gladia

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Gladia connector helper that can manage transcription jobs and audio files through the oo CLI, with notable but purpose-aligned access to sensitive media and deletion actions.

Install only if you trust OOMOL/oo CLI access to your Gladia account and intend to let the agent manage transcription jobs. Confirm uploads, transcription starts, audio downloads, and deletes carefully, especially for confidential recordings, and clean up transit-stored audio when it is no longer needed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (2)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The manifest description says to use this skill for ANY Gladia request and instead of calling the API directly, which is an overly broad trigger that can cause the agent to invoke the skill in many contexts without sufficient task-specific scrutiny. Broad routing language increases the chance that high-impact actions such as uploads, transcription starts, or deletions are selected automatically, especially because the skill exposes both read and destructive operations.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill explicitly states it will download the original audio file and store it in connector transit storage, but it does not warn the user about the privacy and data-handling implications of moving and retaining potentially sensitive audio content. This can lead to unintended collection or persistence of personal, confidential, or regulated data without adequate operator awareness or consent checks.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal