← Back to Skills Marketplace
247
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install my-test-viedo
Description
Extract frames or short clips from videos using ffmpeg.
README (SKILL.md)
Video Frames (ffmpeg)
Extract a single frame from a video, or create quick thumbnails for inspection.
Quick start
First frame:
{baseDir}/scripts/frame.sh /path/to/video.mp4 --out /tmp/frame.jpg
At a timestamp:
{baseDir}/scripts/frame.sh /path/to/video.mp4 --time 00:00:10 --out /tmp/frame-10s.jpg
Notes
- Prefer
--timefor “what is happening around here?”. - Use a
.jpgfor quick share; use.pngfor crisp UI frames.
Usage Guidance
This skill is a thin wrapper around ffmpeg and appears coherent. Before installing, ensure you: (1) install ffmpeg from a trusted package source (brew in the spec) appropriate for your OS; (2) run the script only on video files you trust (it will create/overwrite the specified output path); and (3) be aware brew may not be available on all platforms (Windows). No credentials or network access are requested by the skill.
Capability Analysis
Type: OpenClaw Skill
Name: my-test-viedo
Version: 1.0.0
The skill is a legitimate utility for extracting video frames using ffmpeg. The bash script 'scripts/frame.sh' provides a wrapper for standard ffmpeg commands to capture frames by timestamp or index, and the 'SKILL.md' documentation contains straightforward usage instructions without any signs of prompt injection or malicious intent.
Capability Assessment
Purpose & Capability
Name/description match the implementation: the skill requires the ffmpeg binary, provides a script that extracts frames, and offers a brew install for ffmpeg. All required components are proportional to the stated purpose.
Instruction Scope
SKILL.md instructs running the included scripts/frame.sh with a video path and output path. The script only checks the input file, creates the output directory, calls ffmpeg to extract a frame, and prints the output path — it does not read other files, environment variables, or contact external endpoints.
Install Mechanism
Install spec uses the well-known 'ffmpeg' brew formula to provide the required binary. No downloads from arbitrary URLs or extract-from-remote archives are present.
Credentials
The skill requests no environment variables, credentials, or config paths. The script only uses command-line arguments provided at runtime.
Persistence & Privilege
always is false and the skill does not request elevated or persistent system privileges. It writes only the output file path provided by the caller and does not modify other skills or global agent settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install my-test-viedo - After installation, invoke the skill by name or use
/my-test-viedo - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of the "video-frames" skill.
- Extract individual frames or short clips from videos using ffmpeg.
- Includes scripts for grabbing the first frame or frames at specified timestamps.
- Provides quick tips on output formats and usage preferences.
Metadata
Frequently Asked Questions
What is 视频提取?
Extract frames or short clips from videos using ffmpeg. It is an AI Agent Skill for Claude Code / OpenClaw, with 247 downloads so far.
How do I install 视频提取?
Run "/install my-test-viedo" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 视频提取 free?
Yes, 视频提取 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 视频提取 support?
视频提取 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 视频提取?
It is built and maintained by jili544 (@jili544); the current version is v1.0.0.
More Skills