Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →小红书笔记素材创作技能。当用户需要创建小红书笔记素材时使用这个技能。技能包含：根据用户的需求和提供的资料，撰写小红书笔记内容（标题+正文），生成图片卡片（封面+正文卡片），以及发布小红书笔记。支持 8 种精美排版主题和 4 种智能分页模式。
.claude/skills/itamarzand88-xhs-note-creator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 68% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 53% | 0% |
<!-- source: xhs-note-creator — https://raw.githubusercontent.com/comeonzhj/Auto-Redbook-Skills/main/SKILL.md -->
根据用户提供的资料或需求，创作小红书笔记内容、生成精美图片卡片，并可选择发布到小红书。
> 详细参数文档见 references/params.md
根据用户需求和资料，创作符合小红书风格的内容：
标题：不超过 20 字，吸引眼球，可用数字/疑问句/感叹号增强吸引力。
正文：段落清晰，点缀少量 Emoji（每段 1-2 个），短句短段，结尾附 5-10 个 SEO 标签。
注意：此 Markdown 专为图片渲染设计，禁止直接使用上一步的笔记正文。
文档结构：
markdown--- emoji: "🚀" title: "封面大标题（≤15字）" subtitle: "封面副标题（≤15字）" --- # 正文内容... --- # 第二张卡片内容...（使用 --- 手动分隔时）
分页策略选择：
--- 手动分隔，配合 -m separator-m auto-splitbashpython scripts/render_xhs.py <markdown_file> [options]
默认主题：sketch（手绘素描风格） 默认分页：separator（按 --- 分隔）
常用示例：
bash# 默认（sketch 主题 + 手动分页） python scripts/render_xhs.py content.md # 自动分页（推荐内容长短不定时） python scripts/render_xhs.py content.md -m auto-split # 切换主题 python scripts/render_xhs.py content.md -t playful-geometric -m auto-split # 固定尺寸自动缩放 python scripts/render_xhs.py content.md -m auto-fit
生成结果：cover.png（封面）+ card_1.png、card_2.png...（正文卡片）
可用主题（-t）：sketch、default、playful-geometric、neo-brutalism、botanical、professional、retro、terminal
分页模式（-m）：separator、auto-fit、auto-split、dynamic
> 完整参数说明见 references/params.md
前置条件：配置好 .env 文件中的 XHS_COOKIE（详见 references/params.md）
bash# 默认仅自己可见（推荐先预览确认） python scripts/publish_xhs.py --title "笔记标题" --desc "笔记描述" \ --images cover.png card_1.png card_2.png # 确认无误后公开发布 python scripts/publish_xhs.py --title "笔记标题" --desc "笔记描述" \ --images cover.png card_1.png card_2.png --public
> 默认以「仅自己可见」发布，加 --public 参数才会公开。
scripts/render_xhs.py — 渲染脚本（主推，8 主题 + 4 分页模式）scripts/render_xhs_v2.py — 渲染脚本 V2（备用，7 种渐变色彩风格）scripts/publish_xhs.py — 发布脚本assets/cover.html — 封面 HTML 模板assets/card.html — 正文卡片 HTML 模板assets/styles.css — 公共容器样式assets/themes/ — 各主题 CSS 文件references/params.md — 完整参数参考（主题/模式/发布参数）| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 19,395 | 16,111 | -17% | 1 | 1 | 0% | 3,889 | 4,314 | +11% | 0 | 0 | — |
case-02 | fail→fail | 14,745 | 11,461 | -22% | 1 | 1 | 0% | 2,632 | 3,231 | +23% | 0 | 0 | — |
case-03 | fail→pass | 12,020 | 13,666 | +14% | 1 | 1 | 0% | 2,325 | 3,896 | +68% | 0 | 0 | — |
case-04 | pass→fail | 16,176 | 14,701 | -9% | 1 | 1 | 0% | 3,001 | 3,935 | +31% | 0 | 0 | — |
case-05 | pass→pass | 22,900 | 12,049 | -47% | 1 | 1 | 0% | 3,721 | 3,169 | -15% | 0 | 0 | — |
case-06 | fail→fail | 16,195 | 16,534 | +2% | 1 | 1 | 0% | 3,056 | 4,405 | +44% | 0 | 0 | — |
case-07 | fail→fail | 8,067 | 5,513 | -32% | 1 | 1 | 0% | 1,502 | 2,068 | +38% | 0 | 0 | — |
case-08 | fail→fail | 5,913 | 3,541 | -40% | 1 | 1 | 0% | 1,174 | 1,640 | +40% | 0 | 0 | — |
case-09 | fail→pass | 12,582 | 4,999 | -60% | 1 | 1 | 0% | 2,489 | 1,932 | -22% | 0 | 0 | — |
case-15 | fail→pass | 8,123 | 3,263 | -60% | 1 | 1 | 0% | 1,624 | 1,662 | +2% | 0 | 0 | — |
case-10 | pass→pass | 8,334 | 3,495 | -58% | 1 | 1 | 0% | 1,684 | 1,755 | +4% | 0 | 0 | — |
case-11 | fail→fail | 8,853 | 4,206 | -52% | 1 | 1 | 0% | 1,774 | 1,895 | +7% | 0 | 0 | — |
case-12 | pass→pass | 6,701 | 8,311 | +24% | 1 | 1 | 0% | 1,386 | 2,653 | +91% | 0 | 0 | — |
case-13 | fail→pass | 4,559 | 1,838 | -60% | 1 | 1 | 0% | 890 | 1,359 | +53% | 0 | 0 | — |
case-14 | fail→pass | 9,958 | 3,269 | -67% | 1 | 1 | 0% | 2,233 | 1,641 | -27% | 0 | 0 | — |
case-16 | fail→pass | 15,629 | 2,176 | -86% | 1 | 1 | 0% | 3,591 | 1,463 | -59% | 0 | 0 | — |
case-17 | fail→pass | 8,051 | 2,639 | -67% | 1 | 1 | 0% | 1,625 | 1,507 | -7% | 0 | 0 | — |
case-18 | fail→pass | 8,157 | 2,221 | -73% | 1 | 1 | 0% | 1,589 | 1,433 | -10% | 0 | 0 | — |
case-19 | pass→pass | 11,171 | 3,314 | -70% | 1 | 1 | 0% | 2,147 | 1,722 | -20% | 0 | 0 | — |
case-20 | fail→pass | 8,382 | 2,288 | -73% | 1 | 1 | 0% | 1,591 | 1,451 | -9% | 0 | 0 | — |
case-21 | pass→pass | 4,984 | 2,801 | -44% | 1 | 1 | 0% | 967 | 1,614 | +67% | 0 | 0 | — |
case-22 | pass→pass | 10,956 | 1,513 | -86% | 1 | 1 | 0% | 2,039 | 1,294 | -37% | 0 | 0 | — |
case-23 | fail→fail | 11,213 | 7,527 | -33% | 1 | 1 | 0% | 1,767 | 2,310 | +31% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +39 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.