Whisper STT

PassAudited by ClawScan on May 1, 2026.

Overview

This skill appears to do what it claims: run local Whisper transcription on user-supplied audio files, with only normal setup cautions for installing third-party packages.

This skill looks coherent and purpose-aligned. Before installing, be aware that it requires local Python/ML dependencies and optionally ffmpeg; use a virtual environment if you want to keep it isolated. Only transcribe audio files you are comfortable processing locally and printing to your terminal.

Findings (1)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

Low

#ASI04: Agentic Supply Chain Vulnerabilities

What this means

Installing these dependencies can add or update local software packages before transcription works.

Why it was flagged

The skill asks the user to install third-party dependencies without pinned versions. This is expected for a local Whisper transcription skill, but it relies on external package sources and changes the local environment.

Skill content

pip install openai-whisper torch ... Optional: Install ffmpeg ... brew install ffmpeg ... sudo apt install ffmpeg

Recommendation

Install in a virtual environment where possible, use trusted package managers, and consider pinning versions if reproducibility matters.