ClawHub
Back to skill

Security audit

MH openai-whisper-api

Security checks across malware telemetry and agentic risk

Overview

This skill appears to do what it claims: upload a user-selected audio file to OpenAI Whisper for transcription.

Install only if you are comfortable sending selected audio files to OpenAI for transcription and using an OPENAI_API_KEY. Avoid processing confidential, regulated, or third-party recordings unless that upload is allowed by your policies.

SkillSpector

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
93% confidence
Finding
The script transmits the provided audio file to OpenAI's remote transcription API, but it does not present any explicit warning, consent prompt, or disclosure at runtime that local audio content will leave the host. This can cause unintentional disclosure of sensitive or regulated audio data, especially when the script is used as a reusable agent skill where users may assume processing is local.

External Transmission

Medium
Category
Data Exfiltration
Content
mkdir -p "$(dirname "$out")"

curl -sS https://api.openai.com/v1/audio/transcriptions \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Accept: application/json" \
  -F "file=@${in}" \
Confidence
88% confidence
Finding
This finding is a real data egress path: the script uses curl to send the audio file and optional prompt/language metadata to an external service endpoint. In the context of a transcription skill this transmission is expected functionality, but it still creates privacy and compliance risk if users are not adequately informed or if sensitive recordings are processed without approval.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.