@online4 — criador de Open Skill

@online4

  • nuke_video_generation

    Fastest path from a one-line idea to a full AI video package — design concept sheet, 12-panel beat storyboard, per-shot video prompts with durations, audio plan, burned-in on-screen captions, and a rendered cut. Works in any language the user writes in (English, Hindi, Hinglish, Odia, Bengali, Tamil, Spanish, and others). Use this skill whenever the user wants an AI video, reel, short film, ad, music video, campaign board, storyboard, design sheet, character sheet, frame-by-frame image prompts, shot prompts, or wants to render, generate or produce a video on this platform.

  • cinematic_video_cellcog

    Use when the user wants an HD image or video (text-to-image, image edit, text-to-video, image-to-video, reference-to-video), or a cinematic short or film from an idea, script, song or images where characters, props and locations must stay consistent across shots; also to revise one shot, re-render a clip, or make a new piece in the same style.

  • erotic-video-pro-evp

    A warm, safe-for-work AI companion. Create a character, chat with them, generate tasteful images and short videos, get a respectful Flirt Trainer, and play "Date Night" — a mini dating game where you pick a country and outfit, then every choice moves you up the ladder from Stranger to Lover. No adult content of any kind.

  • ai_video_script_writer

    Write the best possible SCRIPT for AI video generators

  • video_studio_pro

    Unified AI video studio covering design-sheet prompt writing, generator scripts, script-to-video story drama, and longer adult production with face and setting lock. Trigger on video studio, prompt writer, script writer, story drama, Long Film, /film, /h3, EVP, Seedance, Real-Motion, Grok Imagine, or when the user uploads an image, a script/document, or a face lock.

  • atelier_lock

    Use when the user wants to lock one person's likeness (their own face or someone who agreed) or an original character from a reference photo, then make SFW HD stills, 10-second clips or 30/60-second reels that stay on that same person — with 137 looks (anime, film, painting, 3D, Korean, Japanese, after-dark lighting) and 122 motion effects (camera moves, transforms, fashion, atmosphere, film looks); also for face swap, head swap, character swap, look-alike, talking avatar and lip sync — always b

  • recreate_any_video_with_ai

    Use when the user wants to remake, remix, restyle, or match a reference video shot-for-shot (Path A), or replace the actors/subjects in footage with custom character identities from reference photos (Path B), producing continuous generative video prompts and stitch scripts. Also use when the user wants the remake restyled into any trending/current style (cinematic, anime, claymation, VHS, TikTok/FYP, etc.) or wants HD image/video output. The user can choose which video/image/audio models to use.

  • ai_visual_gen_guru

    Visual-first multilingual teacher for AI image and video generation, zero to pro. Every lesson ships a Learning Visual - a ready image/video prompt whose output demonstrates the concept (A/B pairs, process clips, metaphor scenes) - plus an Osho-style discourse mode for ideas that will not land. Covers prompt anatomy, cinematic action prompting, genre prompting (ad, marketing, drama, music video), camera, lens, lighting, composition, editing and control (img2img, inpainting, ControlNet, LoRA, references), character consistency, and the theory behind diffusion and transformer models, seed, CFG, samplers.

  • ugc-product-ad

    Use when the user wants to produce a UGC-style product advertisement video (authentic creator-style ad for a product/SKU) — from product + platform + target length to a finished captioned, scored ad; also when they want to revise only one stage (script, casting, a segment, subtitles, BGM, final mix) of such a production. This is a concept-visualization tool — it produces the visual idea, storyboard, and copy-paste Seedance 2.5 video prompts for the final high-quality footage.

  • action_choreography_consultant

    Research-oriented fight and action choreography consultant that analyzes, edits, and improves fight scenes in any medium, and can visualize them as HD reference stills, storyboard frames, and fight clips. Specialties include Chinese xianxia/wuxia elemental combat (air, water, electricity, wind, fire, earth), anime TV/film fight sequences and sakuga traditions, and live-action martial arts/stunt work. Use when the user provides a script excerpt, scene description, storyboard, beat sheet, or reference clip summary and asks to analyze, critique, improve, redesign, choreograph, or generate visuals for a fight scene

  • motion_control_transfer_sfw

    Copy any motion from an uploaded driving video onto images, whole subject or named parts. Safe for work only. No nudity and no explicit sexual content. Generates the video directly from the best available video model and uploaded references — no preview still is built first. If no image is provided, wait for the user's instruction before proceeding. Model choice follows the if-then table (i2i editor first, then SFW motion-control or i2v). Trigger on motion

  • paper_collage_skill

    Create captivating 5–10 second vintage paper-collage animation videos with handcrafted aesthetic and continuous pop-up book styling, and 3D Breakthrough scenes where a character (or pair) tears out of a flat printed world into real, dimensional space

  • multi-angle-reference

    Use when the user wants to generate multi-angle views of a product, person, model, character, or IP from an uploaded reference image — locking the verified structure/identity, then producing fixed multi-angle grids (e.g. nine-grid line art), commercial/editorial moodboards, single finished shots, expanded angles with original lighting, or 3:4 + 16:9 character sheets; more angles can be added by reusing the anchor image.

  • style-reference-lock

    Use when the user wants to lock a consistent style and/or character identity across AI stills or a sheet (style reference lock, style lock, identity lock, character consistency, match a reference look, reproduce a pose/outfit/scene from an existing pack, or build a reusable style reference pack).

  • depth_map_generator

    Use when the user uploads an image (or describes a scene) and asks for a depth map, depth pass, z-depth, depth estimation, depth-map style visualization, or prep for parallax / 2.5D / 3D-photo / ControlNet-depth / depth-based fog / depth-of-field. Generates a grayscale (or inferno/viridis) depth-style image of the input, near objects bright or dark.