Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    ShortsAI-Studio Skills

    1

    Automate YouTube Shorts creation with AI-driven content strategy and FFmpeg-based video production workflows.

    Free

    Security scanned
    ShortsAI-Studio Skills

    ShortsAI-Studio Skills

    Example session with this skill installed

    See It in Action: From Wildlife Footage to YouTube Short

    Step 1: The Creator's Request

    Prompt

    I have a 34-second wildlife video.

    • 1–5 seconds: A deer runs across the sidewalk toward a tree.
    • 6–12 seconds: It reaches its family, consisting of a mother deer and another fawn.
    • 17–23 seconds: The mother deer stands close to the camera, calmly eating leaves without appearing bothered by the recording.
    • The two young deer move farther into the woods.

    Help me turn this into an engaging YouTube Short. Create a compelling title, opening hook, description, hashtags, and timed on-screen text. Then suggest how to edit the video, including subtitles, music, and pacing.

    Step 2: Content Strategist in Action

    Recommended Title

    She Found Her Family… Mom Wasn't Even Bothered! 🦌

    Alternative Titles

    1. A Deer Family Reunion Right Before My Eyes
    2. Two Fawns Disappear Into the Woods, Mom Just Keeps Eating
    3. A Quiet Moment With a Deer Family

    Opening Hooks

    • "Wait until you see where this deer is going!"
    • "She wasn't alone after all..."
    • "A family reunion in the woods!"

    Description

    A deer suddenly runs across the sidewalk and joins her family. While the two young deer disappear deeper into the woods, Mom calmly enjoys her lunch, completely unfazed by the camera. 🦌

    Hashtags

    #Deer #Wildlife #DeerFamily #Fawns #Nature #WildlifeShorts #AnimalMoments

    On-Screen Text

    TimeText
    0–5 secWhere is she going?
    6–12 secFound her family!
    13–17 secThe little ones head into the woods...
    17–23 secMom? Not even bothered!
    24–34 secJust another peaceful day in the woods

    Step 3: Production Assistant in Action

    Once the creator provides the actual video file, the Production Assistant can help prepare the finished Short.

    Production workflow:

    1. Inspect the original video, duration, resolution, and audio.
    2. Review the footage and identify the key moments.
    3. Generate synchronized subtitles if speech or narration is present.
    4. Add the selected on-screen text at the appropriate timestamps.
    5. Prepare the video in vertical 9:16 format, if needed.
    6. Mix suitable background music while preserving important original audio.
    7. Render the final video using FFmpeg.
    8. Validate the exported video's resolution, duration, audio, and subtitle synchronization.

    Expected deliverables:

    • Finished YouTube Short video
    • Subtitle file (SRT), when applicable
    • Video title and description
    • Relevant hashtags
    • Suggested thumbnail text

    Actual video rendering requires an environment with video file access, FFmpeg, and the necessary processing tools.

    The Result

    Two complementary skills turn a raw video idea into a complete publishing workflow:

    • Content Strategist: What to say and how to capture viewers' attention.
    • Production Assistant: How to turn the footage into a finished video.

    From raw footage to a ready-to-publish YouTube Short.

    • Read your context and instructions
    • Compiled the shortsai-studio

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

    About this skill

    Short-form creators often struggle to bridge the gap between raw footage and a polished, algorithm-ready YouTube Short. This studio package provides a two-stage workflow to handle both creative strategy and technical production using automated video tooling.

    What it does

    • Creative Strategy generates five clickable titles, three opening hooks, and mobile-optimized text overlays based on transcripts or visual analysis.
    • Vertical Reframing converts horizontal source files into 9:16 vertical videos using crop, letterbox, or blur-fill strategies.
    • Automated Subtitling transcribes speech, generates SRT files, and burns synchronized captions directly onto the video.
    • Production Orchestration manages FFmpeg commands for rendering, mixing background music, and applying timed scene overlays.
    • Quality Validation checks output resolution, aspect ratio, and audio levels to ensure the file is ready for upload.

    How it works

    1. Analyze the source material by providing a video file, transcript, or simple description of the content.
    2. Strategize by selecting from generated titles, hooks, and creative directions for the story progression.
    3. Execute production tasks like transcription, music mixing, and vertical rendering using available environment tools.
    4. Verify the final output file for duration, framing, and subtitle synchronization.

    Frameworks & tools

    This skill utilizes FFmpeg and FFprobe for video manipulation, Whisper for speech-to-text transcription, and Python execution environments for file processing.

    Why this beats prompting it yourself

    Writing prompts for video editing often results in generic advice rather than executable commands. This skill structures the metadata for the YouTube algorithm while providing precise FFmpeg parameters and SRT generation logic that standard prompts cannot maintain consistently.

    Use cases

    • Convert horizontal long-form podcast clips into vertical Shorts with burnt-in captions.
    • Generate high-engagement titles and hooks for raw vlog footage.
    • Automate the transcription and subtitling process for non-native speakers.
    • Batch prepare social media content with consistent branding and music levels.

    Known limitations

    Does not include FFmpeg, Whisper models, or Python binaries. Requires an execution environment where these tools are pre-installed to perform rendering or transcription.

    Requires local installation of FFmpeg, FFprobe, and Whisper. Cannot bypass platform copyright checks for music. Resolution limited by source file quality.

    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 1 day ago

    • Free with an Agensi account

    Requires

    Python

    Permissions

    Read Files
    Write Files
    Listed1 day ago

    What's inside

    Frequently Asked Questions

    This skill automates the heavy lifting of short-form production, converting horizontal footage into 9:16 vertical videos with burned-in subtitles, AI-generated hooks, and optimized YouTube titles. It bridges the gap between a raw idea and a render-ready file by managing complex FFmpeg commands and metadata strategy.

    You will receive the structured logic, metadata templates, and precise orchestration parameters required to execute the workflow. Please note that external binaries like FFmpeg, FFprobe, and Whisper must be pre-installed in your agent's execution environment to perform the actual video rendering and transcription.

    The skill offers three distinct strategies: center-cropping for focused subjects, letterboxing to preserve the full frame, or a blur-fill effect that places the source video over a blurred, scaled background for a modern aesthetic.

    Yes, the skill is designed to handle batch processing, allowing you to generate multiple title variations, hooks, and rendered clips for social media content calendars or podcast highlights in a single session.

    By combining Whisper-based transcription with FFmpeg-driven 'hardcoding,' the skill ensures subtitles are perfectly synchronized and rendered directly into the video file, making them visible across all devices without needing external SRT uploads.

    New in Social Media