More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Wiss Shorts

    1

    Spanish-narrated 9:16 shorts with 3D toy characters, live mascots, kinetic headlines and 92 visual styles, rendered on your own computer.

    $20

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Wiss Shorts

    Wiss Shorts

    Example session with this skill installed

    Haz un short sobre la medicina de momias: durante siglos en Europa se vendía polvo de momia como remedio.

    • Read your context and instructions
    • Compiled the wiss shorts

    Summary

    "Medicina de momia": a 30 s looping short narrated in Spanish, built from the script and storyboard you approved and rendered to out/medicina-de-momia.mp4.

    1. Script (approved)

    #LineAccent word
    1La aspirina.aspirina
    2Antes de que existiera, tu médico te podía recetar... polvo de momia.momia
    3Durante siglos, en Europa, eso era medicina.medicina
    4Los boticarios molían momias egipcias y las vendían para el dolor de cabeza, las heridas... para casi todo.todo
    5Y hasta los reyes se la tomaban.reyes
    6Hubo tanta demanda, que las momias de verdad se acabaron.acabaron
    7Así que los traficantes las fabricaban... con muertos de la calle.calle
    8La próxima vez que te duela la cabeza, piensa qué te darían... si no existieraexistiera

    The last line is completed by the first one ("...si no existiera / la aspirina"), so the video loops.

    2. Storyboard (approved)

    • 8 shots, one per line, looks: dither → halftone → risograph → comic → cmyk → ascii → dither, with ascii, blocks, dots, braille, diamond and cross transitions.
    • Mascots snug, mochi and zumi watch the 3D props and react on their word; the first and last shot share the same set so the loop is invisible.

    3. Build and review

    SJ new medicina-de-momia
    SJ voice medicina-de-momia    # 8/8 lines, every word timed
    SJ build medicina-de-momia && SJ audio medicina-de-momia   # music mood: curioso
    SJ preview medicina-de-momia  # out/preview.jpg, all shots scored 8+
    SJ render medicina-de-momia
    

    4. Delivered in out/

    • medicina-de-momia.mp4: 1080×1920, -14 LUFS, plus bucle.mp4 to check the loop
    • portada.png, medicina-de-momia.srt, contact sheet and a 360 px phone check
    • publicar.md: 3 titles, a 2–3 sentence description with the source, 5–8 hashtags and the cover hook

    Next: want another one in the same series, or a different look for this topic?

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

    About this skill

    Give it a topic and get a finished vertical short narrated in Spanish: script, 3D toy scenes, voice, music, effects, subtitles and cover. Everything renders on your computer with HTML and 3D in a headless browser: no GPU, no Blender, no image models and no credits.

    What it does

    • Script with a loop: a hook in line 1, 6–8 lines, 25–40 s, and a last line that the first one completes, so the short plays as a seamless loop. The agent checks the facts and tells you when something was exaggerated.
    • 3D toy characters that act: sets and poses per shot, with the camera always moving and something new every 1–2 s.
    • Live mascots: 14 wissive blobs that breathe, blink, look at what is happening and react on their word.
    • Kinetic headline: word by word, with the key word in italic serif and a hand-drawn underline.
    • 92 visual styles: dither, halftone, risograph, comic, CMYK, ASCII, pixel, glass, glitch and more, plus cell-based transitions between shots.
    • Voice and timing: Kokoro voice per line, with every word timed by Whisper, so every animation is anchored to the narration.
    • Sound: synthesized music (5 moods) and effects, mixed to -14 LUFS.
    • Ready to post: MP4, cover image, .srt subtitles, contact sheet and 3 title options, a description and hashtags.

    How it works

    1. Script: you approve it before anything is built.
    2. Storyboard: one shot per line (set, event, who reacts, style, accent word); you approve it too.
    3. Build: short.json, shots.js and overlay.js, then voice, music and effects.
    4. Review: the agent scores preview frames against a quality checklist and fixes anything under 8/10 before showing you.
    5. Render: about 2–3 minutes per 30 s of video.

    Use cases

    • Curious history or science shorts for TikTok, Reels and YouTube Shorts in Spanish.
    • A recurring series for a Spanish-language channel with one consistent look.
    • Redoing or editing a short you already made with the skill.

    Known limitations

    • Narration is in Spanish (Kokoro voice em_alex).
    • Needs Node.js 22+ and FFmpeg. On first use, with your OK, doctor.mjs creates a private Python environment in ~/.cache/wiss-shorts and downloads the ~340 MB voice model.
    • Reference docs inside the skill are in Spanish.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Video Skills.

    Narration is in Spanish only. Needs Node.js 22+ and FFmpeg; the first run downloads a ~340 MB voice model into a private Python environment. Rendering takes about 2–3 minutes per 30 s. Reference docs inside the skill are in Spanish.

    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 buying a skill.

    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

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 15 days ago

    Software
    Node.js 22+FFmpegPython 3HyperFrames

    tools/doctor.mjs creates a private Python venv. HyperFrames runs with npx.

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Environment Variables
    Network Access

    Runs Node and Python scripts and `npx hyperframes` in your terminal. Writes only to the project folder and to ~/.cache/wiss-shorts (private Python venv and voice model; delete that folder to remove everything). Network: github.com (voice model, once), huggingface.co (Whisper model, first timing run), registry.npmjs.org (HyperFrames), cdn.jsdelivr.net (GSAP, three.js, wissive) and fonts.googleapis.com while rendering. No API keys needed. Your script, voice and video stay on your computer.

    Listed15 days ago

    What's inside

    Frequently Asked Questions

    The narration is in Spanish (Kokoro voice em_alex), and the scripts, headlines and subtitles are written in Spanish. The agent can talk to you in your own language while it works.

    No. Every frame is HTML and three.js rendered by HyperFrames in a headless Chrome on your computer, so it runs on an ordinary laptop with integrated graphics. There are no credits or subscriptions; a 30 s short takes about 2–3 minutes to render.

    You install Node.js 22+ and FFmpeg. On first use the agent asks you first, then `doctor.mjs` creates a private Python environment in ~/.cache/wiss-shorts with kokoro-onnx and faster-whisper, downloads the ~340 MB Kokoro voice model and prepares HyperFrames. Nothing is installed globally; delete ~/.cache/wiss-shorts to remove it all.

    A 1080×1920 MP4 at -14 LUFS, a cover image, .srt subtitles, a contact sheet and a 360 px phone check, a loop check when the short loops, and a ready-to-post text file with 3 titles, a description with the source and 5–8 hashtags.

    Yes. The agent shows you the script and then the storyboard and waits for your OK at each step. You can pick the visual style per shot from 92 looks, the music mood (curioso, misterio, humor, epico, tierno), the mascots and the accent word of each line.

    New in Image & Video