Halloween Makeup Beauty · Open Skill 公開スキル
Halloween Makeup Beauty
Halloween Makeup Beauty は Digen 上の公開 Open Skill です。開くと、このワークフローが何を作るか分かります。
Custom
0 0 0
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4284861/2026-10-15/f9kdfuum.jpg
# Halloween Makeup Beauty Create **one photorealistic Halloween makeup portrait** per request: **face-forward beauty lighting**, crisp makeup detail, and **festive spooky glam**—not cartoon, not body-horror gore. Choose **one of five makeup presets** (or a user custom brief): | # | Preset | |---|--------| | 1 | **Porcelain Doll Cracks** | | 2 | **Candy Skull Floral Crown** | | 3 | **Crystal Spider-Web** | | 4 | **Midnight Witch Glam** | | 5 | **Neon Sugar Phantom** | **Image-to-image** keeps the uploader's **face, identity, skin tone, hair, and visible outfit** while applying makeup and subtle scene mood. **Text-to-image** builds an **original** subject when the user has no photo or taps **direct generate**. Preset prompt blocks and the shared quality lock live in `references/makeup_presets.md`. Channels and assembly live in `references/generation.md`. No project state. **Language lock:** This file and all of `references/` are **English only**. Reply in the **user's language**. Every `generate_image` `prompt` is **English**. Do not tell the user tool names, channel names, preset numbers as internal codes, or dump full makeup prompts. ## User-visible speech (hard) **Allowed lines only:** - Need input: one short note that they may **pick a preset (1–5)**, **upload a clear face photo**, **describe colors and details** (any language in chat), or **tap direct generate** — then pause **P-empty**, **P-preset**, 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 **Halloween makeup portrait** 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 (use **4** on **P-preset** and **P-empty** when listing presets). Localize `text` and `content` to 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 preset, no usable brief, no direct-generate tap, no last image | Direct generate (Preset 1) · Pick another preset · Upload a photo | | P-preset | User wants Halloween makeup but did not name a preset | Four preset shortcuts (see `guidance-buttons.md`) · Custom brief | | P-brief | Text too vague (no subject cues and no preset) | Example brief (`fill_input`) · Direct generate Preset 1 (`send_message`) | | P-refuse | NSFW goal, sexualized minor, gore focus, hate target | Family-safe direct generate · Upload a different photo | | P-fail | `generate_image` `success=false` after retry | Regenerate same preset · Direct generate again · Try another preset | | P-done | One portrait ready | Higher quality · Same preset, new colors · Different preset · Direct generate new look | ## When to use - User wants **Halloween makeup** / **spooky beauty portrait** / **face paint glam** with one of the **five presets**. - User uploads a **clear face photo** and wants **makeup transformation** while **keeping face and outfit**. - User has **no photo** or taps **direct generate** / **generate without upload** — run **text-to-image** immediately (default **Preset 1: Porcelain Doll Cracks** unless they named another preset or tapped a p