← Back to Skills Marketplace
Transcribe
by
javicasper
· GitHub ↗
· v1.0.2
3984
Downloads
2
Stars
39
Active Installs
3
Versions
Install in OpenClaw
/install transcribe
Description
Transcribe audio files to text using local Whisper (Docker). Use when receiving voice messages, audio files (.mp3, .m4a, .ogg, .wav, .webm), or when asked to transcribe audio content.
Usage Guidance
Read the installer before running it. Do not install this version system-wide until the missing scripts/transcribe wrapper is supplied and reviewed, or change the install to a user-local path. If you proceed, expect Docker to fetch external Python/model dependencies during build and verify the language setting before relying on transcripts.
Capability Analysis
Type: OpenClaw Skill
Name: transcribe
Version: 1.0.2
The skill's `scripts/install.sh` uses `sudo` to copy an executable file named `transcribe` to `/usr/local/bin`. The content of this `scripts/transcribe` file is not provided in the analysis input. Installing an unknown executable with elevated privileges to a system-wide path is a risky capability, as its actual behavior cannot be assessed, leading to a 'suspicious' classification despite the benign nature of the other provided files and stated purpose.
Capability Assessment
Purpose & Capability
The visible artifacts align with local Whisper transcription in Docker and do not show credential access, exfiltration, destructive behavior, or unrelated data collection.
Instruction Scope
Runtime instructions are limited to saving an audio attachment, running a transcribe command, returning text, and cleaning up the temp file; the documented Spanish default can affect transcription accuracy but is disclosed and overrideable.
Install Mechanism
SKILL.md claims a scripts/transcribe wrapper exists, and install.sh uses sudo to copy it to /usr/local/bin/transcribe, but the package contains only SKILL.md and scripts/install.sh. That missing executable makes the install path incomplete and leaves the intended system-wide wrapper unavailable for review.
Credentials
Docker and faster-whisper are proportionate for local transcription, but the build pulls unpinned external dependencies and a model at install time, so the resulting image can vary over time.
Persistence & Privilege
The installer creates a persistent Docker image and attempts a privileged system-wide CLI install under /usr/local/bin. No background service or credential persistence is shown, but the privileged mutation is material given the missing wrapper.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install transcribe - After installation, invoke the skill by name or use
/transcribe - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
Dockerfile embedded in install.sh - fully self-contained
v1.0.1
Fix: include docker folder with Dockerfile
v1.0.0
Initial release: local Whisper transcription via Docker
Metadata
Frequently Asked Questions
What is Transcribe?
Transcribe audio files to text using local Whisper (Docker). Use when receiving voice messages, audio files (.mp3, .m4a, .ogg, .wav, .webm), or when asked to transcribe audio content. It is an AI Agent Skill for Claude Code / OpenClaw, with 3984 downloads so far.
How do I install Transcribe?
Run "/install transcribe" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Transcribe free?
Yes, Transcribe is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Transcribe support?
Transcribe is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Transcribe?
It is built and maintained by javicasper (@javicasper); the current version is v1.0.2.
More Skills