Guide · Agent skill
How to use the Video Frame Extractor agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The video-frames skill lets an AI agent pull a single still image out of a video file, using the free ffmpeg tool to do the work. It is aimed at people who want a quick thumbnail, a preview of a moment in a clip, or a screenshot of a UI moment from a screen recording. By the end of this guide you will know what the skill does, how to install it, how to ask for a frame in plain language, and what to check before you use it.
Video Frame Extractorspinabot/brigade · MIT Download the skill Quick start
- 1Install ffmpeg. Install the ffmpeg command-line tool, for example through the Homebrew formula named ffmpeg, and confirm it works on your machine.
- 2Download the skill ZIP. Get the video-frames ZIP from this page.
- 3Unzip into a skills folder. Unzip it into ~/.claude/skills/ for personal use or into .claude/skills/ inside a project.
- 4Or upload it in the apps. In the Claude apps, go to Settings, then Capabilities, then Skills, and upload the skill.
- 5Ask for a frame. Name the video file path and say whether you want the first frame or a timestamp, and whether you want JPG or PNG.
What is an agent skill, and what does this one do?
An agent skill is a folder of instructions and, sometimes, scripts that an AI agent can load when a task matches the skill. The skill is described in a SKILL.md file, and this one also includes a shell script, scripts/frame.sh.
Once installed, the agent can extract a single frame from a video file. It can take the first frame, a frame at a timestamp such as 00:00:10, or a frame by its index number, and save it as a JPG or PNG. The script runs ffmpeg with the options the agent chooses.
- Output a single frame per run, saved as .jpg or .png
- Choose the first frame, a timestamp, or a frame index
- Useful for thumbnails, previews and checking a specific moment
Install the video-frames skill
Download the ZIP of the skill from this page. Then install it in one of two ways. For a personal setup, unzip it into ~/.claude/skills/. For a single project, unzip it into that project's .claude/skills/ folder.
Alternatively, upload the skill in the Claude apps under Settings, then Capabilities, then Skills.
- Personal: unzip into ~/.claude/skills/
- Project: unzip into .claude/skills/ inside the project folder
- Claude apps: Settings → Capabilities → Skills, then upload
Install ffmpeg, which the skill requires
The skill will not work without ffmpeg, a command-line tool for processing video and audio. The skill metadata suggests installing it with Homebrew, using the formula named ffmpeg. The skill's homepage is https://ffmpeg.org.
If ffmpeg cannot be installed on your machine, this skill will not function there.
- Required tool: ffmpeg
- Suggested install route in the skill metadata: Homebrew formula ffmpeg
How do you trigger the skill with a natural request?
You do not need special syntax. Describe the image you want in plain language and include the path to the video file. The agent decides when the skill fits the task.
Ask for the first frame when you want an overview. Ask for a timestamp when you want to know what is happening around a specific moment. The skill's notes recommend --time for questions like what is happening around here.
- Example: Grab the first frame of /path/to/video.mp4 as a JPG.
- Example: Get a frame at 00:00:10 from /path/to/video.mp4 and save it as a PNG for a crisp UI screenshot.
- Example: Make a quick thumbnail from this screen recording.
Run the script directly: the commands in the skill
The skill's quick start shows two patterns. The first takes the first frame and writes it to /tmp/frame.jpg. The second takes a frame at 00:00:10 and writes it to /tmp/frame-10s.jpg. The {baseDir} placeholder points to the skill's own folder.
Use .jpg for quick sharing and .png for crisp UI frames.
- First frame: --out sets the output file
- Timestamp: --time sets the moment, in HH:MM:SS form
{baseDir}/scripts/frame.sh /path/to/video.mp4 --time 00:00:10 --out /tmp/frame-10s.jpgSafety notes: scripts, file access and network
The skill runs a shell script, so review scripts/frame.sh before you install it if you are unsure what it does. The agent is given direct access to local file paths, so only point it at videos you have the right to process. Avoid sensitive recordings you do not want the agent to read.
The skill metadata states that it does not use the network.
- Runs scripts: yes (scripts/frame.sh)
- Uses network: no
- Gives the agent access to local file paths you name
Limits and tips
The script saves one frame per run. It does not cut clips and is not built for bulk extraction. For many frames, you would need to run it repeatedly or use a different tool.
If a frame looks wrong, try a different timestamp. Keep in mind that the skill is a quick inspection tool, not an editor.
- One frame per run; no clip cutting in the script
- Not designed for extracting many frames at once
- Choose .png for UI detail, .jpg for smaller shareable files
What people use it for
- Making a quick thumbnail from a video for a post or listing.
- Capturing a UI moment from a screen recording to share with a team or include in documentation.
- Previewing what happens at a specific timestamp in a clip without opening an editor.
Frequently asked questions
Is the video frame extractor skill free?
The skill is published under the MIT licence. It relies on ffmpeg, which the skill describes as a free tool. The source does not list any pricing for the skill itself.
What do I need to use this skill?
You need ffmpeg installed on your machine, because the skill will not function without it. The skill metadata suggests installing it with Homebrew using the formula named ffmpeg.
Where do I put the skill after downloading it?
Unzip it into ~/.claude/skills/ for personal use, or into .claude/skills/ in a project folder. You can also upload it in the Claude apps under Settings, then Capabilities, then Skills.
Can it extract a whole video or many frames at once?
No. The script saves a single frame per run and does not cut clips. For bulk extraction you would need to run it repeatedly or use another tool.
Does the skill send my videos over the internet?
The skill metadata states that it does not use the network. It does run a shell script and gives the agent access to local file paths, so review the script and choose videos carefully.
Should I use JPG or PNG for frames?
Use JPG for quick sharing. The skill notes suggest PNG for crisp UI frames, where fine detail such as text matters.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you need a quick still image from a video, such as a thumbnail, a preview of a moment, or a UI screenshot from a screen recording, and you can install ffmpeg. Pick something else if you need to extract many frames or cut clips in bulk, or if you cannot install ffmpeg or do not want an agent reading local video files.
Download the skill