Bathroom Vanity Selfie Pose · عوامی Open Skill
Bathroom Vanity Selfie Pose
Bathroom Vanity Selfie Pose Digen پر ایک عوامی Open Skill ہے۔ کھولیں اور دیکھیں یہ ورک فلو کیا تخلیق کرتا ہے۔
Product
4 0 0
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4284862/2026-10-16/4j6v99wk.jpg
# Bathroom Vanity Selfie Pose Photo-recapture skill: recast one user-supplied reference image into a realistic, stylish bathroom-vanity mirror selfie. The person must stay recognizable; only the setting, pose and camera distance are rebuilt around them. Output is a still image. ## When to use - The user uploads an image and wants a bathroom-vanity mirror selfie made from it. - The user asks to "redraw / recapture / restyle" a photo into that mirror-selfie pose, or wants more pose options for the same photo. - The user wants a cleaner, more polished version of the same shot for a profile picture or portfolio. - Not for: editing a photo without changing the scene, generating a person from text only, or producing video (this skill returns a single image). ## Steps 1. **Confirm the input.** This skill needs one reference image. If the user has not uploaded a person photo, ask them to attach one and stop — never invent or describe a person that was not provided. 2. **Read the visual conventions before generating.** Call `read_skill_file` with `name="bathroom-vanity-selfie"` and `path="references/style-and-prompt.md"`. That file holds the style, content, layout, must-keep and must-avoid rules plus the English prompt template. Follow it; do not improvise a rival style. 3. **Read the step checklist when the input is unusual** (several people in frame, phone already visible, unclear identity, very low resolution, or a first attempt that came back distorted): `read_skill_file` with `path="references/step-details.md"` for the intake checklist, prompt assembly order and retry criteria. 4. **Assemble the English prompt** following `references/style-and-prompt.md`: identity traits taken from the reference image first, then bathroom vanity setting, then pose, then camera distance and framing, then lighting/materials/color, then quality tail. Every generation prompt is English. 5. **Generate**, one image per call: `generate_image(prompt=<assembled English prompt>, input_urls=[<reference image>], model='i2i.hd.lite', orientation='portrait', resolution='1080P', name='Bathroom Vanity Selfie', entity_type='selfie_images')` - The reference image goes in `input_urls` (first entry = the person to preserve). - `model` is the image-to-image channel; keep it as `i2i.hd.lite` for this recapture look. Do not pass a physical model name. - `orientation='portrait'` and `resolution='1080P'` are the defaults for this theme; if the platform clamps them for the channel, accept the returned adjustment instead of retrying with bigger values. 6. **Check the result** against the must-keep / must-avoid list in `references/style-and-prompt.md` (identity recognizable, face fully visible, hands and phone sane, no extra people or text). If it fails, re-generate at most once or twice with a corrected prompt rather than looping. 7. **Reply in one or two natural sentences**: state that the mirror selfie is ready and what was preserved (identity, hairstyle, outfit if visible). Do not paste long URLs — the asset is already on the canvas. ## Notes - Preserve identity, body proportions and any detail the user explicitly asked to keep; never change gender, age, ethnicity, hairstyle or glasses. - Do not add anything the application does not provide: no text captions, watermarks, logos, stickers, app UI, or extra people. - Image-only skill. If the user asks for a video of the same scene, say this skill produces a still image and offer another pose instead. - If `generate_image` returns `success=false`, tell the user plainly what failed; never pretend it worked or fabricate an image link. - Keep prompts and content SFW. Politely decline sexual, violent or otherwise policy-violating requests without calling the tool (mirror-selfie generation uses the `i2i.hd.lite` channel, which is not available for NSFW content).