@pendenskull — Open Skill 創作者

@pendenskull

• storytelling

  Write long-form stories paragraph-by-paragraph from an overview, continue them on request, steer with "what happens next" ideas, keep memory of long stories via rolling summaries, and generate scene images for the story. Use when the user asks to write, continue, or plan a story, novel, creepypasta, fantasy, anime, mystery, or any fiction, or wants a scene image for the current story.

• shotcaller_v1_full_cinema_director

  Wan 3.0 Director — writes production-grade video prompts for Alibaba Wan 3.0 (text-to-video, first/last frame, reference-to-video, video editing, extensions). Sets duration and resolution first (2–30s, 480p–1080p), maps references as Image 1 / Video 1 / Audio 1, then builds a shot brief: style, references, critical blocks, geometry, first frame, lens, camera, light, timestamped action, physics, acting, audio and locks. Handles dialogue from a pasted script (lines spoken verbatim, lip-synced from

• image-outpainting-output

  A production-oriented skill for turning an existing image into multiple aspect-ratio outputs using image outpainting. The skill prioritizes output quality, composition, product/subject fidelity, technical validation, safe text zones, and clean handoff rather than generation alone.

• music_video_creator_low_credits

  Turn an uploaded song into a 16-9 music video. Trigger on song to music video, automatic music video, full-song video, 4-minute music video, 24-shot storyboard, animate the track, or when the user uploads a song and wants a complete visual video. Present style presets and allow a custom style. Use Grok Imagine stills plus ffmpeg because generate_video and mix_clips are not connected here.

• sketch_to_video_ai_workflow

  Turn a sketch, drawing, storyboard, line-art frame, diagram, or hand-drawn artwork into an AI-generated video while preserving the intended visual medium, composition, silhouette, topology, and line language. Use when the user wants sketch-to-video, drawing animation, animate-my-sketch, image-to-video from line art, storyboard-to-video, or a sketch-based motion workflow.

• furry_character_art_skill

  Create recognizable, consistent furry and anthropomorphic characters by defining visible identity features, choosing the right generation/editing route, translating prompts to the selected model, and validating the entire character rather than judging only the face or rendering quality.