Style Transformation 2 · Open Skill 공개 스킬
Style Transformation 2
Style Transformation 2은 Digen의 공개 Open Skill입니다. 열어 보면 이 워크플로가 무엇을 만드는지 알 수 있습니다.
Custom
1 2 5
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4277547/2026-10-14/aggyju2r.jpg
# Style Transformation 2 Dress an uploaded **adult woman or female character** in the **Style Transformation 2** Lovely Bunny cosplay set from preset `style2_outfit_reference`: **white satin** bra-style top with **black trim** and thin straps, **black satin** high-cut bottom with **side tie strings**, **black headband** with **upright white fluffy bunny ears**, **white fabric collar** with **black bow tie**, and **white fluffy wrist cuffs**. Structure and accessories follow the preset when bound. No project state; subject images must be `asset_id` or conversation URI already in the thread. Anime and real photos share the same image-to-image path. Do not mention physical model names, channel names, or full prompts to the user. English prompts and channels live in `references/outfit-swap.md`. **Language lock:** This file and all of `references/` are **English only** (no Han / CJK characters). Reply in the **user's language**. Every `generate_image` `prompt` is **English**. ## User-visible speech (hard) Do not narrate the pipeline. Never mention skill files, `read_skill_file`, silent redo, or channels. **Allowed lines only:** - Need input: one short question; always stress **upload a full-body photo for best results** when asking for a photo. Then pause **P-empty** or **P-brief**. - Starting generate: at most **one** short line (user language), then tool calls only — **no** `set_guidance` on that turn. - Success: one or two sentences that the cosplay look is ready — no URL, no prompt dump. Pause **P-done**. - Tool failure: brief honest note. Pause **P-fail**. - Safety block: one polite refusal. Pause **P-refuse**. ## Pause = clickable buttons (hard) Any turn that **stops without a new shown image**, and **every success turn**, must end with: 1. One or two short sentences (user language). 2. **`set_guidance`** in the **same turn**, with **at least two** clickable options. **No tool calls after `set_guidance`.** Do **not** call `set_guidance` in the same turn as `generate_image` or `recognize_image`. Each button is an object with `text` and `action` (not bare strings): - **`send_message`**: `content` is the full sentence sent when tapped. - **`fill_input`**: `content` is starter text the user can edit (`replace: true` when replacing the input box). Offer **2–4** options. Translate `text` and `content` into the user's language. No channel or tool names on buttons. Before every pause turn, `read_skill_file` → `references/guidance-buttons.md`. | Pause | When | Example intents (localize) | |---|---|---| | P-empty | No upload, no usable character description | I'll upload a full-body photo · Upper-body only · Text-only female character | | P-brief | Text too vague (no subject or framing) | Example brief (`fill_input`) · I'll upload a full-body photo instead | | P-accessories | Upload OK but user did not say ears/collar/cuffs | Full Lovely Bunny set default · Ears and outfit only · Minimal accessories | | P-refuse | Minor, full nudity goal, sexualized minor, male subject | Upload adult female full-body photo · Family-safe outfit instead | | P-fail | `generate_image` `success=false` after retry | Regenerate same photo · Upload a different full-body photo | | P-done | Shown transformation ready | Regenerate · Softer lighting · New full-body photo · Adjust accessories | ## When to use - User wants **Style Transformation 2** / **Lovely Bunny** cosplay try-on on an uploaded **woman or female character**. - User asks to toggle **accessories** (ears, collar, bow tie, cuffs) or **regenerate** the last result. - Text-only: user describes an adult female character (hidden base, then one shown swap). **Do not use:** male subjects, minors, objects without a person, generic retouch with no outfit change, background-only edits, video, unrelated garments, or **Style Transformation** (v1 pink lace set) unless the user clearly wants this v2 preset. ## Preset reference asset After publish, bind the cover flat-lay as preset **`style2_o