Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    video loop

    1

    Turn any video into a seamless perpetual loop locally using specialized ffmpeg blending and cut-point detection scripts.

    Free

    Security scanned
    video loop

    video loop

    Example session with this skill installed

    I have a 5-second AI video of drifting clouds (clouds.mp4). I need it to loop seamlessly for a website background. Can you help me pick a method and run it?

    • Read your context and instructions
    • Compiled the video loop

    I will analyze clouds.mp4 for a clean cut point first. Since clouds are diffuse textures, if a clean cut isn't found, I will use the half-period blend method which is superior for hiding seams in organic motion.

    Loop Analysis and Command

    MetricValue
    Source Fileclouds.mp4
    Resolution1920x1080
    Best MSE Ratio2.4x (Threshold >10x)
    Recommended Methodhalfperiod-loop.sh
    Predicted Seam SSIM0.942
    
    # Setting script path (local project)
    S=.claude/skills/video-loop/scripts
    
    # Executing half-period blend with a 2x slowdown to maintain duration
    # This blends the first half with the second for a seamless texture
    bash "$S/halfperiod-loop.sh" clouds.mp4 clouds_loop.mp4 2
    
    # Verifying frame count and seam integrity
    ffprobe -v error -count_frames -select_streams v:0 -show_entries stream=nb_read_frames -of csv=p=0 clouds_loop.mp4
    

    Next steps

    • Run ffplay -loop 0 clouds_loop.mp4 to visually inspect the seam.
    • Update your project manifest with the new metadata (Method: halfperiod, Cost: 0).
    • If the overlay is too visible, try the seamless-loop.sh script with a 10-frame crossfade.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Creating seamless video loops is difficult because standard crossfades often produce "ghosting" or visible seams that break the illusion. This skill provides a set of specialized local scripts to convert existing video files—whether AI-generated or filmed—into perpetual loops using deterministic ffmpeg operations. It moves beyond simple fades by offering multiple looping strategies based on the specific motion characteristics of your source material.

    What it does

    • Automatic cut-point detection analyzes frames to find identical states for perfect, zero-blend cuts.
    • Half-period blending overlays the two halves of a clip to eliminate localized transitions entirely.
    • Localized crossfading fades the end of a clip into the start for short-duration motion.
    • Motion interpolation uses slowdown parameters to lengthen loops without increasing overlay artifacts.
    • SSIM quality measurement calculates the mathematical seam quality to predict visibility before rendering.

    How it works

    1. Run the find-loop-point.py script to check if the video has periodic motion suitable for a clean cut.
    2. Select a blending script based on the visual content: halfperiod-loop.sh for diffuse textures or seamless-loop.sh for short, low-amplitude motion.
    3. Apply an optional slowdown factor to achieve the target duration while maintaining high SSIM values.
    4. Verify the loop integrity using ffprobe to ensure no frames were dropped during the concat process.

    Frameworks & tools

    This skill requires ffmpeg, ffprobe, bash, and python3. It relies on the standard Python library and local shell execution, ensuring all processing stays on your machine with no API costs.

    Why this beats prompting it yourself

    Writing ffmpeg filter graphs for loops is error-prone, especially regarding N variable indexing and timestamp drift in concat filters. These scripts handle the non-obvious math of linear ramps and frame normalization that manual prompts often miss, preventing the "one-frame jump" common in basic loops.

    Use cases

    • UI Backgrounds creating infinite hero video backgrounds for web apps.
    • AI Post-processing turning one-shot AI video generations into loopable social assets.
    • Digital Signage preparing silent, perpetual motion content for displays.
    • Live Wallpapers generating high-quality animated desktop backgrounds.

    Known limitations

    All audio tracks are automatically dropped. Works best on slow, continuous motion; turbulent flows like smoke or flames may show ghosting if SSIM values are below 0.75.

    Drops all audio tracks. Requires local ffmpeg/ffprobe installation. Blending quality depends on source motion amplitude.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 22 days ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed22 days ago

    What's inside

    Frequently Asked Questions

    This skill replaces manual frame-matching and basic crossfading with mathematical cut-point detection and specialized blending strategies, eliminating the common 'one-frame jump' or ghosting artifacts found in basic loops.

    The purchase includes a suite of Python and Bash scripts, including the loop-point detection tool, multiple blending scripts for different motion types, and a quality verification utility.

    You will need local installations of FFmpeg, FFprobe, Bash, and Python 3; all processing is performed locally on your hardware to ensure privacy and eliminate recurring API costs.

    Yes, this skill is designed to handle both standard filmed footage and AI-generated clips, though it is particularly effective at turning short AI generations into long-duration, high-quality loops.

    While this skill can be integrated into most AI agent workflows that support shell execution, it is optimized for agents that can interact with local file systems and run CLI-based scripts.

    The scripts are optimized for visual continuity, so all audio tracks are automatically stripped during the looping process to ensure a perfectly seamless visual transition.

    New in Workflow Automation