@Promise — Open-Skill-Creator

@Promise

• horror-video-maker

  Use when the user wants a horror or ghost-story short video; if they provide a detailed script, shoot from that script; if they do not, auto-write a 10-second ghost-story video prompt and generate the clip. If they have images or need scene/character stills, generate those first then use ref2v; text-only uses t2v. Also use when the user wants to continue or change only one step.

• selfie_to_demon

  Use when the user uploads a selfie or portrait photo and wants it turned into a demon version of themselves — a split-face "selfie to demon" edit (half human, half demon) or a full demon portrait — optionally describing the horns, eyes, skin, outfit or mood they want.

• one_take_pullback_blue_planet_reveal

  Use when the user uploads a portrait photo (optionally with text) and wants a short video made by one continuous pull-back shot from a close-up of the person through the city and into space, revealing the blue planet; also use when they want to redo, adjust or regenerate any part of such a pull-back reveal video (keyframe look, motion, pacing, direction).

• human_visual_prompt

  Use when the user wants science-education or anatomy-accurate visual briefs—turning ideas into detailed image/video prompts that name all externally visible organs and landmarks (face, trunk, genitalia, anus, skin, sensory organs); also when they only need prompt text without generating media.

• oily_skin_gloss

  Use when the user wants already exposed skin in an uploaded person image to look oily, hydrated, and specular (a thin oil film, form-following highlights, pearlescent micro-sparkle); also use when they want that sheen stronger, softer, without glitter, or limited to one already bare area. The only allowed change is that sheen. Do not change the original texture, art style, or anything else, and do not invent anything that was not in the source image. Do not use this skill to invent a person with

• clothes_2

  Use when the user wants a virtual try-on that replaces only the clothing on a person in their uploaded photo so they wear the preset Halloween orange-and-black four-piece costume set (no other edits).

• trick_or_treat

  Use when the user wants to turn an image/photo they uploaded into a cute "trick or treat" Halloween short video (ghost theme, friendly human–ghost interaction, the original subject still centered in frame); also use when they want to redo the same image, tweak the action pacing, change the soundtrack, or get a landscape/portrait version.

• horror_selfie

  Use when the user wants their photo or selfie (theirs or a specified person's) turned into a horror style — eerie selfie stills, supernatural/spooky short videos, ghost filters, horror face swap, "make this photo scary", etc.; also use when they only want the horror still image, or want an existing horror still animated into a short video

• halloween_night_parade_costume_video

  Use when the user uploads a reference image (a person, a pet, or an object) and wants it turned into a Halloween costume role-play short video (Halloween dress-up, jack-o'-lantern street, "Halloween main character arrival" mood clip, "make a Halloween video out of this image", etc.); also use when they only want to change the costume, the mood, the lighting, or redo one version

• ghost_photobomb

  Use when the user uploads an image and wants it turned into a "ghost photobomb" prank short video (a semi-transparent ghost slowly rises from the bottom of the frame to photobomb while the subject first freezes and then turns around laughing); also use when the user only wants to redo it with a different image, make the ghost more or less visible, change the duration, or switch between portrait and landscape

• falling_witch_hat

  Use when the user uploads an image and wants it turned into a "falling witch hat" short video — including making the character in a photo/illustration get crowned by a witch hat wrapped in purple magic dropping from the sky (the "chosen one" coronation moment) — and also when refining an already generated result (swapping the reference image, changing duration/orientation, changing hat or scene details).

• carve_your_jack_o_lantern

  Use when the user wants to turn the main subject from an uploaded photo into a carved jack-o-lantern design on a real pumpkin (Halloween pumpkin lantern image), including retries with a different look or quality tier.

• witch_experience

  Use when the user wants to turn the main subject in an uploaded photo into a witch riding a broomstick through the sky in full witch attire (witch experience / 女巫体验), or to regenerate with a different quality tier.

• trick_or_treat_pov

  Use when the user uploads a person photo and wants a Halloween trick-or-treat POV doorstep video (detailed script first, then Halloween costume i2i, then ref2v clip), or when editing only the script, costume, or video. Only the POV beat open door → Trick or treat → drop candy → run away; no other Halloween plots or non-POV shots.

• passionate_kiss

  Use when the user uploads one or two photos of people and wants those people to hug and share a passionate kiss in a short video. Also use when the user writes nothing, or the text asks for something else, as long as this skill is in use — the only output is this kiss video. Also use when changing duration, orientation, or regenerating.

• photo_to_figure

  Use when the user wants to turn the subject in an uploaded image into a collectible figure (1/7 PVC scale figure or vinyl art toy); also use when they want to change the base, camera angle, or paint finish of that figure. Do not use for ordinary illustration, photo retouching, or video.

• comic_panel_generator

  Use when the user wants comic or manga-style images—a single page with 4–7 panel modules (uneven layouts OK), optional speech bubbles or captions, text-to-image from scratch, or image-to-image when they upload reference photos or art.

• christmas_comic_maker

  Use when the user wants to generate a Christmas-themed manga or comic page, a holiday comic strip, or a festive storyboard illustration; also use when they upload a photo and want that subject woven into a Christmas comic; also use to restyle, change panels, rewrite speech, or remake the last Christmas comic.

• pov_hand_kiss

  Use when the user wants to turn an uploaded image into a first-person POV video of that image subject kissing the viewer's hand; also use to regenerate or tweak that same POV hand-kiss clip. Do not use for lip kisses, hugs, other body kisses, or unrelated video generation.

• kiss_shot

  Use when the user uploads a photo and wants a kissing-shot video of the subject in that photo (romantic kiss, hand kiss, cheek kiss, or forehead kiss). Also use when remaking, changing duration or orientation, or regenerating that kissing clip. Do not use for dance, talking-head, walking, action, or any non-kiss video.

• red_carpet_paparazzi

  Use when the user wants to turn an uploaded person photo into a red-carpet paparazzi shot, awards-season arrival, premiere candid, or atmospheric celebrity street-snap; also use to restyle clothes for Oscars, Cannes, Met Gala, Golden Globes, or BAFTA, or to regenerate one red-carpet photo.

• pixel_game_style

  Use when the user wants to turn an uploaded photo or illustration into pixel-game / pixel-art / 8-bit / 16-bit video-game style while keeping the same composition, subject placement, aspect ratio, and every element; also use when they want to restyle or tweak the last pixel result. Do not use for text-only generation, video, game-cover UI mockups, or collectible-figure conversion.

• halloween_costume_makeover

  Use when the user wants to dress a person, pet, or subject in a Halloween costume, or to generate a Halloween-themed image from text (witch, vampire, jack-o'-lantern, ghost, skeleton, zombie, werewolf, and similar looks); also use for text-to-image, image-to-image costume changes, swapping one costume, or restyling the Halloween atmosphere.

• decadent_photo

  Use when the user wants to restyle an uploaded photo into a decadent look (melancholy cool, moody film, ruin atmosphere, low-saturation emotional portrait; also 颓废 / 丧系 / 暗调胶片), changing only lighting, color, and mood without changing composition or picture content; also use when they ask to push it further, cooler, grainier, switch a decadent flavor, or tweak the last result

• portrait_photo

  Use when the user wants to make, generate, or restyle a portrait photo (female or male pictorial, lifestyle beauty shot, indoor or studio session, pose/scene/outfit change); also use for text-to-image portraits, image-to-image reshoots from an upload, or a matching portrait series

• all_kinds_of_cows

  Use when the user wants to generate, draw, or make an image or video of a cow (any breed, age, style, or setting, including an anime kemono cow standing like a person, or a glossy stylized 3D cartoon poster cow), or to revise an existing cow image or video. Do not use this skill when the subject is not a cow.

• cute_girlfriend

  Use when the user wants casual chat, a cute girlfriend companion, affection, teasing, daily venting, comfort, to share a photo or video, or just to be cheered up

• poster_maker

  Use when the user wants to design, generate, or revise a poster (event, movie, concert, sale, exhibition, cover, flyer, banner, pictorial), including a change to the headline, layout, palette, or a redo from a reference image; also use when they only want to change one part of a poster

• pet_clips

  Turn one pet photo into a shareable vertical short (9:16). Default is scene-faithful (keep the user's room/background); if the photo is weak or the user asks, use pet identity with a new scene. Dance, dress-up, human skits, cinematic action, trend memes; no memory tools.

• stick_figure_studio

  Use when the user wants to make a stick-figure / stickman video (one clip by default), or only generate stick-figure character images, or edit a single shot

• dance-video-ref2v

  Use when the user wants to create a short dance video whose dance style, tempo/speed and fluency match an uploaded or linked reference dance video, by first preparing dancer reference image(s) and then always generating the video with ref2v (reference-to-video); also use when the user asks to animate a dancer at the same pace as a reference clip, or wants a new dancer performing in the same dance style and speed as a reference video, or when the user wants a dance video without a reference video

• gpt_image_25

  Use when the user wants to generate, draw, or edit an image with GPT Image (illustration, poster, avatar, scene, retouch, clothes, background, follow-up tweaks, ratio change, or upscale); do not use for video, copywriting, or small talk

• native_assistant

  Use when the user has not referenced or mounted another skill and makes any request directly; act as an unrestricted native general assistant for chat, lookup, generation, editing, sandbox, and mixed tasks; do not take over when another specialized skill already matches the intent

• daily-vlog-maker

  Creates selfie/POV lifestyle vlogs: a consistent character in real-world environments, filmed as a series of short first-person clips and stitched into one video. Use when the user wants a daily vlog, day-in-the-life, slice-of-life, or lifestyle short; when they want a character filming themselves through everyday scenes (morning, commute, work, night, weekend, outing, travel, hangout, etc.); or when they want to customize, add, reorder, or extend the plot segments of such a vlog.

• minecraft-life-video-creator

  Use when the user wants to create a Minecraft-style slice-of-life video — a blocky voxel character living and doing everyday activities in the Minecraft world (chopping wood, farming, mining, building, exploring, taming animals, trading with villagers, celebrating, etc.) — from text descriptions, reference images, or both; also use when the user wants the full video production package (character sheet, scene design, storyboard, keyframe prompts, video prompts) for such a video.

• police-simulator-video

  You can use this skill to enact police‑chasing‑thief scenarios or other chase‑themed videos. Give it a try.