← Back to Skills Marketplace
kartinw

Video-Watch

by kartinW · GitHub ↗ · v1.0.0
cross-platform ✓ Security Clean
3546
Downloads
6
Stars
37
Active Installs
1
Versions
Install in OpenClaw
/install video-watcher
Description
Analyze video content by extracting frames at regular intervals. Use when you need to understand what's in a video file, review video content, analyze scenes, or describe video without being able to play it directly. Supports MP4, MOV, AVI, MKV, and other common video formats.
Usage Guidance
Install only if you are comfortable with videos being converted into still image files in the workspace. Use a dedicated output folder for sensitive videos, avoid reusing folders with existing frame_*.jpg files you need to keep, clean up extracted frames afterward, and install ffmpeg only from trusted package managers.
Capability Analysis
Type: OpenClaw Skill Name: video-watcher Version: 1.0.0 The skill bundle is benign. The `SKILL.md` clearly describes the purpose of analyzing video content by extracting frames and provides instructions for the AI agent to use the `scripts/extract_frames.sh` script. The script itself uses standard tools (`ffmpeg`, `ffprobe`) to perform the stated task, handling video input and output directories as expected. There is no evidence of prompt injection attempts against the agent, data exfiltration, malicious execution, persistence mechanisms, or obfuscation. All actions are directly related to the stated purpose of video frame extraction.
Capability Assessment
Purpose & Capability
The stated purpose is video analysis by frame extraction, and the script only validates a user-provided video path, reads metadata with ffprobe, and extracts JPEG frames with ffmpeg.
Instruction Scope
The workflow asks the agent to run a local extraction script and inspect generated images; this is purpose-aligned, but it means video contents are converted into separate files the agent can view.
Install Mechanism
The skill requires ffmpeg and suggests apt or brew installation. That dependency is relevant and disclosed, but users should use trusted package sources.
Credentials
The script creates an output directory and writes frame_*.jpg files there. It uses ffmpeg overwrite mode, so existing matching frame files in the selected output directory can be replaced.
Persistence & Privilege
There is no evidence of background execution, privilege escalation, credential use, or exfiltration. Extracted frames remain on disk until deleted.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install video-watcher
  3. After installation, invoke the skill by name or use /video-watcher
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Enable your Clawbot to watch video
Metadata
Slug video-watcher
Version 1.0.0
License
All-time Installs 37
Active Installs 37
Total Versions 1
Frequently Asked Questions

What is Video-Watch?

Analyze video content by extracting frames at regular intervals. Use when you need to understand what's in a video file, review video content, analyze scenes, or describe video without being able to play it directly. Supports MP4, MOV, AVI, MKV, and other common video formats. It is an AI Agent Skill for Claude Code / OpenClaw, with 3546 downloads so far.

How do I install Video-Watch?

Run "/install video-watcher" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Video-Watch free?

Yes, Video-Watch is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Video-Watch support?

Video-Watch is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Video-Watch?

It is built and maintained by kartinW (@kartinw); the current version is v1.0.0.

💬 Comments