Snow Hooded Mage Orb · Open Skill 公開スキル

Snow Hooded Mage Orb

Snow Hooded Mage Orb 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。

@jimi

Custom

0 0 1

https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/3325120/2026-10-13/hw5v43yz.webp

# Snow Hooded Mage Orb

Turn an **adult male** face reference and/or a **written character brief** into one **hyper-realistic cinematic still**: a hooded mage **standing in falling snow**, **cold blue ambient light**, **soft warm golden light on both hands** cradling a **bright glowing orb**. The **dark-fantasy snow set**, **orb-holding pose**, and **film-grade rendering** stay locked; only identity (optional photo), **robe trim / ornaments**, and **facial runes** vary from user text.

No structured project state. Progress is conversation + canvas asset.

## Global language lock

Handbook (this file + all `references/`) is **English only—zero Han/CJK**. User-facing copy uses locked **`user_lang`** per `references/00-language.md`. Image `prompt` is **always English**.

## Pause + guidance (hard)

**Every time you stop and wait for the user**, call **`set_guidance` in the same turn** as the last user-visible message—**no exceptions**. Pauses include: missing photo/text, multiple faces, policy refusal, explain-the-look without generate, recoverable failure, and **after every successful image**.

Before any pause turn, `read_skill_file` → `references/guidance-buttons.md` and pick **2–4** actionable buttons for that situation.

- **Forbidden:** ending a wait turn with only prose and no `set_guidance`.
- On the `set_guidance` turn: **no other tools** in that turn.

## User-visible speech (hard)

The user must not see the pipeline. Tools still run; chat text does not narrate them.

**Never tell the user:** that you are reading a skill or reference; todos; tool names; channel names (`t2i`, `i2i.hd.lite`, …); `SCENE LOCK`; the full prompt; `entity_type`; silent QC or redo.

**Allowed user-visible text:**

- Need input: ask for an **adult male** photo and/or a short character description (outfit ornaments, runes). **Stop** with `set_guidance` per guidance-buttons.
- Starting: at most one short line in **`user_lang`**. Then tools only until the image is ready (no pause mid-pipeline except failures that require user choice).
- Success: one or two sentences in **`user_lang`**. Same turn: **`set_guidance` only** (no other tools).
- `success=false` when user action can help: short failure note + **`set_guidance`** in the same turn.
- Refuse: short polite refusal + **`set_guidance`** (allowed alternatives).

## When to use

- User wants a **realistic / cinematic / dark fantasy** mage in **snow** with a **glowing orb**, from a **male portrait** or **text concept**.
- User wants to **change robe decoration**, **facial runes**, or **swap face reference** while keeping the **same snow scene, lighting, and orb pose**.
- User asks to **regenerate** or **refine** the last result with the same locked scene.

Do **not** use for: video-only requests, unrelated illustration styles, **female / child / ambiguous minor** subjects, NSFW, or scenes without the hooded snow mage + orb concept.

## Reference files

| File | When |
|---|---|
| `references/scene-template.md` | Assemble prompt, redo, or user asks what the look is |
| `references/guidance-buttons.md` | Any pause turn (before `set_guidance`) |
| `references/00-language.md` | First user-visible message; before `set_guidance` |

Do not tell the user you are reading these files.

## Channel

Every `generate_image` must pass `model`. No physical model names. Never bare `i2i`.

| Mode | When | `input_urls` | `model` |
|---|---|---|---|
| **i2i** | User uploaded a portrait (identity reference) | required (that image, or last result if editing) | default **`i2i.hd.lite`**; user asks HD / best quality → **`i2i.hd`**; user names GPT image → **`i2i.hd.gpt`** |
| **t2i** | No usable face image; adult male described in text only | omit | default **`t2i`**; user asks HD → **`t2i.hd`** |

Main path: **optional photo + text → i2i** when a portrait exists; **text-only → t2i** when the user has no photo or insists on pure text.

## Steps

All internal unless noted in “User-visible speech”.

1. **Intake**
   - Lock *