Voice To Protocol Transcriber

ReviewAudited by ClawScan on May 10, 2026.

Overview

The skill mostly matches its lab-note purpose, but its save-file handling can write outside the intended folder and its setup/privacy details need review.

Review the save-location behavior before using this skill, especially if experiment names may come from copied text or voice-derived content. Pin and verify dependencies, clarify whether any speech recognition is network-based, and store generated protocol files in an approved secure folder.

Findings (4)

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.

What this means

A malformed or untrusted experiment name could make the skill save files somewhere the user did not expect.

Why it was flagged

The experiment name is used directly in the output path with only spaces replaced. If it contains '../' or starts with an absolute path, the saved protocol can be written outside the configured save directory.

Skill content
filename_base = f"{self.experiment_name.replace(' ', '_')}_{timestamp}" ... file_path = self.save_directory / f"{filename_base}.md"
Recommendation

Sanitize experiment names, strip path separators, resolve the final path, and reject saves unless the resolved file remains inside the configured save directory.

What this means

Users may install different package versions than the author tested, or may not know which dependencies are actually required.

Why it was flagged

The skill documents manual installation of unpinned packages, while the registry has no install spec and requirements.txt does not list these same dependencies.

Skill content
pip install speechrecognition pyaudio pydub python-docx
Recommendation

Provide a proper install specification or pinned requirements file that matches the documented dependencies.

What this means

A user could assume recordings stay local even though the documentation suggests some recognition modes may need network access.

Why it was flagged

The documentation gives mixed signals about whether speech recognition uses a network service, which matters for audio that may contain lab or clinical details.

Skill content
"Chinese recognition requires good network connection" ... "Network Access | No external API calls | Low"
Recommendation

Clarify whether recognition is fully local or uses external services, and require explicit user approval before sending audio or transcripts to any provider.

What this means

Saved protocol files may contain sensitive research or clinical information and will remain on disk unless the user manages them.

Why it was flagged

The skill is designed to persist experiment records locally, and the documented use cases include clinical experiment logging.

Skill content
"save_directory": "~/Documents/Experiment-Protocols" ... "Clinical experiment operation logging"
Recommendation

Store outputs only in approved locations, apply appropriate access controls, and avoid recording regulated or confidential data unless storage is compliant.