菜单
NovelAI image generation for RP. Extracts [img:...] prompts from story text and generates images via NovelAI API. Supports V4.5/V4/V3 all models. Use when the user asks to generate images, /img commands, or wants to create illustrations for RP scenes.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
AI roleplay sessions using a SillyTavern-compatible runtime engine. Use when the user asks for AIRP, SillyTavern-style play, sends preset/character/world-book JSON, asks for character cards, personas, world books, Tavern Card V2, group chats, swipes, continuations, or a complete RP pack.
Create, edit, and list SillyTavern World Book (世界书) entries via CLI. Use when the user wants to: (1) create a new world book JSON file, (2) add character/worldbuilding/rule entries to an existing world book, (3) edit specific entries by UID, (4) delete entries, (5) list entries, (6) batch-modify positions or trigger keys for multiple entries. Supports 8 injection positions (↑Char ↓Char ↑AT ↓AT @D ↑EM ↓EM Outlet), trigger keywords, depth, constant/selective activation, and more.
Create, edit, and list SillyTavern World Book (世界书) entries via CLI. Use when the user wants to: (1) create a new world book JSON file, (2) add character/worldbuilding/rule entries to an existing world book, (3) edit specific entries by UID, (4) delete entries, (5) list entries, (6) batch-modify positions or trigger keys for multiple entries. Supports 8 injection positions (↑Char ↓Char ↑AT ↓AT @D ↑EM ↓EM Outlet), trigger keywords, depth, constant/selective activation, and more.
| name | novelai-image-gen |
| description | NovelAI image generation for RP. Extracts [img:...] prompts from story text and generates images via NovelAI API. Supports V4.5/V4/V3 all models. Use when the user asks to generate images, /img commands, or wants to create illustrations for RP scenes. |
当 RP 过程中需要生成图片时，按以下流程操作：
/img 生成 → 从最近的 [img: ...] 提取提示词并生成/img 横版 → 同上，1216x832 横版/img furry → 切换 Furry 模型/img 模式 N → 用指定模型号生成/gen <提示词> → 直接生成# 从故事文件提取 + 生成
python scripts/extract-img.py rp-output.txt -g
# 直接生成
python scripts/novelai-generate.py -p "提示词"
# 横版
python scripts/novelai-generate.py -s 1216x832 -p "提示词"
# Furry
python scripts/novelai-generate.py --furry -p "wolf girl, ..."
# 纯场景
python scripts/novelai-generate.py --background -p "sunset, ..."
所有命令在 生图skills/ 目录下执行。
| 模型 | API ID |
|---|---|
| V4.5 Curated ★ | nai-diffusion-4-5-curated |
| V4.5 Full | nai-diffusion-4-5-full |
| V4 Curated | nai-diffusion-4-curated-preview |
| V4 Full | nai-diffusion-4-full |
| V3 Anime | nai-diffusion-3 |
| V3 Furry | nai-diffusion-furry-3 |
标签不是脑补的，而是从 RP 正文中提取:
原文: 苏晚晴解开围裙，蹲下身握住那根巨大的肉棒，
龟头已渗出透明前列腺液。
提取 → [img: 1boy, 1woman, mature woman squatting, holding big penis,
precum dripping, morning light, nsfw, sex, masterpiece, best quality]
提取要素: 谁(人数/年龄) → 外貌 → 衣着 → 动作/体位 → 表情 → 环境 → 画质
在 生图skills/.env 中设置:
NOVELAI_API_KEY=你的API密钥
references/prompt-guide.md — 提示词编写指南references/params-guide.md — V3/V4 参数参考