← Back to Skills Marketplace
Video Transcribe - 视频转文字
by
aaa2531349
· GitHub ↗
· v1.0.3
· MIT-0
606
Downloads
0
Stars
4
Active Installs
3
Versions
Install in OpenClaw
/install video-transcribe
Description
本地视频转文字 - 使用 OpenAI Whisper 进行语音识别，完全免费、离线运行、保护隐私
Capability Analysis
Type: OpenClaw Skill
Name: video-transcribe
Version: 1.0.3
The skill bundle is classified as suspicious primarily because it performs automatic, unprompted installation of external software. The `transcribe.py` script uses `subprocess` to execute `pip install` with the `--break-system-packages` flag to download and install the `openai-whisper` package (approx. 300MB) from a remote mirror. While this behavior is disclosed in `SKILL.md` and `PUBLISH.md` as a convenience feature, the execution of package managers to fetch and install external code at runtime is a high-risk capability. Additionally, `transcribe.py` contains hardcoded local developer paths (e.g., `/Users/seven/...`), which, while not malicious, indicates poor environment handling.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install video-transcribe - After installation, invoke the skill by name or use
/video-transcribe - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.3
**v1.0.3 Changelog**
- 首次运行时自动安装 Whisper 依赖，无需手动执行 pip 命令
- 安装依赖时自动使用清华镜像源，加快国内用户安装速度
- 文档（SKILL.md）更新，提示用户自动安装行为
- 其余功能保持不变，优化用户首次使用体验
v1.0.2
- Added YAML front matter to SKILL.md to fix compatibility with OpenClaw.
- Updated metadata files for ClawHub specification (_meta.json added, skill.json removed).
v1.0.0
- Initial release of Video Transcribe (版本 1.0.0)
- Transcribe local video/audio to text with OpenAI Whisper, fully offline and free
- Supports multiple formats: mp4, mov, avi, mkv, mp3, wav, m4a, and more
- Auto language detection (90+ languages), SRT subtitle export, and multiple model choices
- Generates AI-powered summaries and key points after transcription (optional)
- All processing is local, protecting user privacy
Metadata
Frequently Asked Questions
What is Video Transcribe - 视频转文字?
本地视频转文字 - 使用 OpenAI Whisper 进行语音识别，完全免费、离线运行、保护隐私. It is an AI Agent Skill for Claude Code / OpenClaw, with 606 downloads so far.
How do I install Video Transcribe - 视频转文字?
Run "/install video-transcribe" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Video Transcribe - 视频转文字 free?
Yes, Video Transcribe - 视频转文字 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Video Transcribe - 视频转文字 support?
Video Transcribe - 视频转文字 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Video Transcribe - 视频转文字?
It is built and maintained by aaa2531349 (@aaa2531349); the current version is v1.0.3.
More Skills