Communitygithub.com

NAI5 社群提示詞工程

模型適配器,將完成的角色或插圖藍圖編譯為 NovelAI V5 社群格式提示詞——包含加權畫師堆疊、全域風格層、場景基礎、角色 1/角色 2 區塊、source#/target#/mutual# 互動標籤,以及使用 weight::tag:: 語法的目標負面引導。用於用戶要求 NAI5、NovelAI、NAI提示詞、標籤提示、Danbooru 風格提示時。不負責設計——若無藍圖,應先經由 aesthetic-director-core 與 character-design-engine / illustration-direction。

NAI5 社群提示詞工程 是什麼?

此技能是專為 NovelAI V5 設計的社群格式提示詞工程適配器。它的核心功能是將已完成的角色或插圖藍圖(blueprint)轉換為結構化的 NAI5 社群提示詞,包括加權畫師堆疊、全域風格層、場景基礎、角色區塊以及角色間互動標籤。此技能適用於 Claude Code 等 AI 代理,主要處理用戶明確要求 NAI5、NovelAI 或 Danbooru 風格提示詞的場景。它不會憑空設計角色或場景——若無預先存在的藍圖,會先引導至其他專注於美學指導與角色/插圖設計的技能。技能文件詳細說明了格式架構,例如畫師分配原則、風格與角色資料分離、角色區塊的標籤排列規則(如身分 token 優先、因果關係透過複合標籤保留),以及如何透過加權與負面標籤精確控制生成結果。文件還強調了避免常見陷阱,如防止被刪除的元素以標籤形式回流,或刺點顏色重複出現。整體而言,這是一個高度專業化的翻譯工具,旨在將精確的設計意圖轉譯為 NAI 模型能理解的標籤語言。

相容平台~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/Tera-Dark/personal-agent-skills/tree/HEAD/02_creation/nai5-community-prompt-engineering

在你喜歡的 AI 中提問

開啟一個已預先載入此 Agent Skill 的新對話。

說明文件

NAI5 社群提示詞工程 是做什麼的?

1. 定位

本 Skill 是適配器:把已經做完設計決定的 blueprint 翻譯成 NovelAI V5 社群格式。

v2.0.0 起合併了原 skills/nai5-prompt-engineering(v2.0.0)與 02_creation/nai5-community-prompt-engineering(v1)兩個副本。原副本裡的 "Personal Aesthetic Rules" 段已刪除——品味只在 personal-identity-profile,本 Skill 不維護副本。

輸入檢查:同 anima-prompt-compiler。沒有命題 / 輪廓 / 四層服裝 / 刺點 / 鎖定事實的輸入不是 blueprint,退回 router。

2. 社群格式架構

[Artist Stack]

[Global Style Layer]

[Scene Base]

char1:
...

char2:
...

Negative:
...

詳見 references/community-format.md。

3. Artist Stack

0.6::artist:name::
  • 每位畫師承擔一個明確職責:線條 / 上色 / 構圖 / 角色設計感——不重複職責
  • 常見 5–12 位,權重 0.25–1.5
  • 必要時控制 -1::artist collaboration::
  • 畫師選擇服務 blueprint 的視覺語言,不替代設計。見 references/artist-stack.md

4. Global Style Layer

風格與角色資料分離。見 references/style-layer.md。

關於品質詞的模型特例:personal-identity-profile/references/workflow-style.md 說"避免通用品質詞"——那是針對 Anima 等模型。NovelAI 的品質 / 美學標籤是訓練過的有效 token,在 NAI 上保留:quality(如 masterpiece, best quality, very aesthetic)、complexity、rendering 三組各取所需,不堆疊。這是適配器層面的合法例外,不是對上游規則的違反。

5. Character Block

char1:
girl, [identity], [hair], [eyes], [expression], [outfit base→structural→extension→accessory], [props], [pose/action]
  • 身份 token 靠前;簽名特徵先於次要裝飾
  • 一個 block 一個人;兩人不混塊
  • blueprint 的因果和密度要落成 tag 的順序和取捨:主錨點相關 tag 靠前且完整,安靜區只給一個大塊面 tag(如 plain ankle-length robe),不給它加裝飾 tag
  • 髮型寫形狀分佈,不寫眾數(asymmetrical hair, long hair on one side, undercut on the other 而不是 long hair)
  • 見 references/character-block.md、references/tag-taxonomy.md

6. Interaction Tags

多角色用 source# / target# / mutual#,動作緊跟所屬角色。見 references/interaction-tags.md。

7. Scene Block

角色情緒與動作先於背景裝飾。背景只保留能解釋光或動作的一層。見 references/scene-block.md。

8. Weighting & Negative

1.5::tag::    -2::tag::
  • 只給:畫師混合、關鍵風格方向、關鍵角色特徵、不想要的風格抑制
  • 不給每個 token 都加權
  • Negative 解決具體問題(風格抑制、偽影),不複製大段通用負面表。見 references/weighting.md、references/negative-strategy.md

9. 翻譯 blueprint 時的取捨

Tag 格式天然會丟失"關係"。補救:

  • 因果鏈 → 用複合 tag 保留最關鍵的一環(sleeve slipped to elbow, weight on left leg)
  • 密度圖 → 用 tag 數量分佈表達:密集區 6–10 個 tag,安靜區 1–2 個
  • 刺點 → 顏色詞只出現一次,緊貼它的落點名詞(red thread on spool),其它地方不再出現該顏色
  • 留下的怪 → 必須進 char block,且靠前;它是記憶點
  • 刪掉的東西 → 確認沒有以 tag 形式偷偷回來(尤其 jewelry, earrings, belt, petals, sparkles)

10. 輸出

用戶要 NAI5 時預設輸出社群格式程式碼區塊。不用 MJ 式長自然語言,除非用戶要。

11. 反饋輪

  • 偽影 / 串線 / 權重問題 → 本 Skill 內單變數調整
  • 設計問題(太平淡、太亂、不像)→ 退回 evaluation-loop → aesthetic-director-core/references/feedback-diagnosis.md

12. 輸出前檢查

  • 輸入是 blueprint?
  • 畫師棧每位有職責、無重複?
  • char block 順序反映了主錨點 / 安靜區?
  • 刺點顏色只出現一次?
  • 被刪掉的物件沒有以 tag 回流?
  • 權重只用在必要處?
  • 鎖定事實未改?

References

  • references/community-format.md — 整體格式
  • references/artist-stack.md — 畫師棧工程
  • references/style-layer.md — 全域風格層(quality / complexity / rendering / 風格抑制)
  • references/character-block.md — 角色塊順序與規則
  • references/tag-taxonomy.md — tag 分類與服裝描述法
  • references/interaction-tags.md — 多角色互動
  • references/scene-block.md — 場景塊
  • references/weighting.md — 權重系統
  • references/negative-strategy.md — 負面策略

Individual skills in this repo

This repo contains 13 individual skills — each has its own dedicated page.

Tera-Dark/personal-agent-skills

Creative direction layer that turns a vague request into one committed design idea before any blueprint or prompt is written. Runs a sequence of generative "moves" (find the obsession, plant a contradiction, pick from the tail, build causality, subtract, keep one strange thing) and produces a short Creative Brief with rejected alternatives. Use for any OC / character / illustration / fashion / key-visual request, whenever output feels generic, "AI-flavored", too plain, too busy, or when the user asks for taste, direction, 审美, 创意方向, 人味, 高级感, 不要AI味.

Tera-Dark/personal-agent-skills

Persistent identity and taste layer for Tera-Dark's creative work. Holds the aesthetic signature (restrained base, one strange thing, a hint of danger, fashion-grade garment construction, female-oriented OC/gacha sensibility), hard dislikes (watches, cyber/mech, random butterflies-roses-particles, adjective costumes), business goals and workflow style. Does not generate prompts. Load first for any creative, design, illustration, prompt or aesthetic-judgment task; also when the user says 我的风格, 我喜欢, 按我习惯, 个人偏好.

Tera-Dark/personal-agent-skills

Entry point for all creative requests in this skill hub. Classifies intent (OC/character design, illustration, fashion, NAI5 prompt, Anima prompt, image reverse analysis, prompt review, ComfyUI/LoRA/dataset), loads identity + aesthetic direction first, then hands off to the right specialist and model adapter. Use whenever a request involves 设计, OC, 人设, 立绘, 插画, 服装, 提示词, prompt, NAI, NovelAI, Anima, 反推, 分析图片, ComfyUI, LoRA, or when it is unclear which skill should handle a creative task.

Tera-Dark/personal-agent-skills

Model adapter that compiles a finished character or illustration blueprint into Anima-ready English prompts in a disciplined Tag block + Natural Language block format. Handles Anima-specific format contract, length budgets, positive-first output, V1 faithful / V2 enhanced modes, model profiles and artifact troubleshooting. Use when the user asks for Anima prompts, Anima 提示词, or names an Anima checkpoint. Does not design — if no blueprint exists, route through aesthetic-director-core and character-design-engine / illustration-direction first.

Tera-Dark/personal-agent-skills

Turns a Creative Brief (from aesthetic-director-core) into a complete, model-agnostic character blueprint — design thesis, silhouette architecture, visual anchors, garment engineering (base / structural / signature extension / accessory system), material contrast, palette hierarchy, behavioral pose + camera, narrative residue, presentation format — then runs a subtraction pass. Use for OC, 人设, 角色设计, 服装设计, 立绘, 高定服设, 二游角色, character sheet, fashion concept. Never writes model-specific prompt syntax.

Tera-Dark/personal-agent-skills

Model adapter that compiles a finished character or illustration blueprint into a natural-language image prompt for models that are not Anima or NAI5 — Midjourney, DALL-E / GPT Image, Google Imagen / Gemini image, Flux, SDXL-style checkpoints, or an unnamed target. Keeps parameters out of the prompt unless the target officially supports them, and labels every model-specific claim with an evidence level. Use when the user names any other image model, says 通用提示词, MJ, Midjourney, DALL-E, Imagen, Flux, SD, 或没说用什么模型. Does not design — if no blueprint exists, route through aesthetic-director-core first.

Tera-Dark/personal-agent-skills

Turns a Creative Brief into an authored, model-agnostic illustration blueprint — the image thesis, captured moment, visual motif, environment-character relationship, camera and framing, scale contrast, negative space, physical light sources, density map, narrative residue, and one coherent surreal detail. Use for 插画, 氛围图, 竖屏, 半留白, 印象风, 故事感, key visual, poster, scene illustration, cinematic composition, or whenever a character needs to be placed into a designed moment rather than displayed on a plate. Never writes model-specific prompt syntax.

Tera-Dark/personal-agent-skills

Decomposes a reference image into transferable design language — silhouette and mass distribution, motif grammar, garment construction, material contrast, palette hierarchy, pose logic, negative space and presentation format — instead of a tag dump. Output feeds aesthetic-director-core (for original work) or a model adapter (for faithful reproduction). Use for 反推, 分析图片, 提取提示词, 还原风格, 参考这张图, reverse prompt, image analysis.

Tera-Dark/personal-agent-skills

Reviews an existing image-generation prompt by separating design intent from model syntax — what it is trying to say, which design layer is weak (thesis, silhouette, garment structure, causality, density, punctum), what is noise, and what a rewrite should change. Use for 优化提示词, 这个 prompt 哪里有问题, prompt review, 分析这段提示词, why does this prompt not work.

Tera-Dark/personal-agent-skills

Technical placeholder for ComfyUI workflow organization, node troubleshooting, metadata and LoRA/model integration. Use when the user mentions ComfyUI, workflow json, nodes, 工作流. Does not handle design or prompt creativity.

Tera-Dark/personal-agent-skills

Technical placeholder for image dataset organization, caption/tag management and training preparation. Use when the user mentions dataset, 数据集, 打标, captions, 训练集. Does not handle visual design decisions.

Tera-Dark/personal-agent-skills

Technical placeholder for LoRA training pipelines: dataset prep, caption organization, training configuration, evaluation. Use when the user mentions LoRA, 训练, fine-tune, 炼丹. Does not decide character concepts or artistic direction.

Tera-Dark/personal-agent-skills

Evaluates a generated image or compiled prompt against the original Creative Brief and blueprint, using a fixed rubric (identity preservation, outfit binding, spatial clarity, unrequested additions, V1/V2 consistency, output contract) plus a design-layer read (thesis, silhouette, causality, density, punctum, one strange thing). Diagnoses which layer failed and routes the fix to that layer with a single-variable change. Use on any feedback round: 太平淡, 太乱, 不像, 没人味, 这版可以, 评价一下, 哪里不对, compare versions.

相關技能