Whisper GPU Audio Transcriber
Security checks across malware telemetry and agentic risk
Overview
This skill appears to match its stated purpose: it runs a local Whisper transcription script, with normal caution needed for Python dependencies, model downloads, and generated subtitle files.
Before installing, make sure you are comfortable with Python package installation, a large first-run Whisper model download, and local creation or overwriting of .srt subtitle files. The provided code and documentation are coherent with the stated transcription purpose.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
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.
The first use may download a large model and requires trusting the Python packages and model source.
The skill depends on external Python packages and downloads a Whisper model on first use. This is expected for local Whisper transcription, but it means installation and first run involve third-party downloads and local disk use.
openai-whisper: Automatically installed via `pip install openai-whisper`; First run will auto-download the model file (turbo ~1.5GB); Models cache in `~/.cache/whisper`
Install dependencies from trusted package indexes, verify the PyTorch/Whisper variants match your hardware, and expect the first run to use network bandwidth and disk space.
Using the skill runs local Python code on the selected audio file.
The skill explicitly instructs the agent to run a local Python script. The provided script aligns with the transcription purpose, so this is expected rather than suspicious.
AI will execute the `scripts/transcribe.py` script
Review the included script if you are cautious about local code execution, and run it only on files and directories you intend to process.
A subtitle file will be created or replaced in the audio file directory or chosen output directory.
The script creates the requested output directory if needed and writes an SRT file. This is necessary for the skill, but it may overwrite an existing same-named .srt file in the target directory.
os.makedirs(output_dir, exist_ok=True) ... output_file = os.path.join(output_dir, f"{basename}.srt") ... open(output_file, "w", encoding="utf-8")Use a dedicated output folder or back up any existing same-named SRT files if overwriting would matter.