Smart Outfit Swap Photo · compétence Open Skill
Smart Outfit Swap Photo
Smart Outfit Swap Photo est une Open Skill publique sur Digen. Ouvrez-la pour voir ce que ce workflow crée.
Custom
0 1 1
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4284864/2026-10-20/yqa9aqp4.jpg
# Smart Outfit Swap Photo Produce **exactly one** shown **image** in which the subject from an **uploaded photo** wears a **new outfit** while everything else stays identical: same face and identity, same hair, same pose and body position, same lighting, same background, same framing. Only the garment changes — with accurate fabric, folds, color, texture, and shadows that sit naturally on the body. This is a photorealistic fashion-editing result for advertising design and role-play, not a stylized illustration. See `references/outfit-swap-contract.md`. **Fixed play:** one shown **image** per request. **Never call `generate_video` or any video tool.** No project state. **Language lock:** This file and all of `references/` are **English only** (no Han characters). Reply in the **user's language**. Every `generate_image` `prompt` is **English**. Do not tell the user tool names, channel names, or the full prompt. ## 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 note that they may **upload a photo**, **describe the outfit**, or **tap direct generate** — 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 **outfit-swapped image** 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. 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 outfit brief, no direct-generate tap, no last success to refine | Direct generate now · Describe the outfit · I'll upload a photo | | P-brief | Upload present but outfit vague, or text-only with no garment | Tailored black suit (`fill_input`) · Summer linen dress (`fill_input`) · Direct generate default (`send_message`) | | P-refuse | NSFW goal, gore focus, sexualized minor, identity/face swap of a real person | Family-safe outfit swap · Upload a different photo | | P-fail | `generate_image` `success=false` after retry | Regenerate same photo · Try a simpler outfit · Upload a different photo | | P-done | One outfit-swapped image ready | Change outfit again · Stronger fabric detail · Keep pose, new background · Direct generate new look | ## When to use - User wants to **change the clothing** on a person in a photo while keeping **pose, face, and background** the same. - User **uploaded a photo** (even with no text) or opened this skill from the catalog — **run `generate_image` immediately** with a sensible default outfit if none is given; do not stop on **P-empty**. - User taps **direct generate**, **generate now**, or similar — **run text-to-image immediately** with the default scene in `references/guidance-buttons.md` unless they already gave a different brief. - Follow-up on the **last successful image** from this skill: swap to another outfit, push fabric detail, or adjust framing (still **one** image). **Do not use:** **video** or motion clips; **face or identity swaps**; **pose chang