DeepSeek Video Studio

    1

    Turn long-form footage into polished 30-second shorts with smart cuts, language-detected captions, reframing, transitions, and QC—without generating media.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    DeepSeek Video Studio

    Example session with this skill installed

    Turn this 45-minute interview into five engaging 30-second vertical clips. Detect the spoken language, remove dead air, add accurate synchronized captions, and export MP4 plus SRT files.

    • Read your context and instructions
    • Compiled the deepseek video studio

    A ranked shortlist and edit plan, followed by five verified 9:16 MP4 clips with source-language captions, SRT sidecars, and a QC report. No generated video or audio.

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

    About this skill

    A DeepSeek editing harness for transcript-led short-form video post-production. It identifies compelling moments, builds an edit plan, removes dead air, reframes for vertical platforms, generates synchronized captions with automatic spoken-language detection, renders with FFmpeg, and verifies the result. Existing media only: no video, audio, voice, music, or B-roll generation.

    Actual rendering requires a DeepSeek agent environment with local file and shell access. Automatic subject tracking depends on optional local vision tools. Transcription accuracy varies with audio quality, accents, overlapping speakers, and code-switching. Thirty seconds is a target, not a forced duration.

    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 5 days ago

    • Free with an Agensi account

    Requires

    FFmpeg
    ffprobe
    Python 3.10+

    Permissions

    Terminal / Shell

    File Scopes

    User-selected
    source
    video/audio
    files
    Skill
    working
    directory
    and
    export

    Designed for local processing. No external media uploads or generative media tools are required. Source files are read-only; exports are written to a separate folder.

    Listed5 days ago

    What's inside

    Frequently Asked Questions

    No. It edits existing footage and audio only. It does not generate video, voice, music, sound effects, or B-roll.

    Yes, through a locally available Whisper-compatible transcription engine. It preserves the spoken language and can handle code-switching when the ASR engine supports it.

    No. Thirty seconds is the target. It may use a 20–40 second range to preserve a complete thought and natural ending.

    It attempts tracking only when a suitable local tool is installed and tested. Otherwise it uses a deliberate crop and checks framing.

    A DeepSeek agent with local file and terminal access, FFmpeg, ffprobe, and Python. A local Whisper-compatible engine is recommended for transcription.

    New in Workflow Automation