Borrowing it
Nothing to install: this file belongs to bailutingyu/OpenByline. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/bailutingyu/OpenByline/main/.claude/skills/article-to-xhs-cards/SKILL.mdgit clone --depth 1 https://github.com/bailutingyu/OpenBylineWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/bailutingyu/openbyline/article-to-xhs-cards)<a href="https://agentmods.dev/skills/bailutingyu/openbyline/article-to-xhs-cards"><img src="https://agentmods.dev/badge/skills/bailutingyu/openbyline/article-to-xhs-cards/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/bailutingyu/openbyline/article-to-xhs-cards"><img src="https://agentmods.dev/badge/skills/bailutingyu/openbyline/article-to-xhs-cards.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00124 | $0.03524 |
| Opus 5 | $0.00062 | $0.01762 |
| Sonnet 5 | $0.00025 | $0.00705 |
| Haiku 4.5 | $0.00012 | $0.00352 |
Grade A, and why
article-to-xhs-cards scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
文章转小红书 / 小绿书图文卡片（风格选择器版）
把一篇文章变成一叠竖图发小红书 / 小绿书。三种画风、3:4 竖版、你自己的 IP 形象。
信条：一图一个认知锚点；主角承担画面核心动作，不是站旁边装饰；为这篇内容重新发明一个怪诞但成立的物理隐喻，不套模板、不做 PPT。
〇、启动先选风格（每次必做 ← 本技能的入口动作）
不要默认画风。 进入任务先把下面 3 种风格的"一句话气质 + 适用场景"列给用户，让用户挑一种（或指定某几张用 A、某几张用 C）。挑定后再进流程，并加载对应 styles/ 文件。
| 风格 | 气质 | 最适合 | 细则 |
|---|---|---|---|
| A 暖彩动漫 | 暖·有人味·故事感 | 封面首图、情绪共鸣、产品故事 | styles/warm-anime.md |
| B 留白混血 | 高级·克制·留白（含彩色档 / 黑剪影档） | 职场共鸣、观点隐喻、正文配图 | styles/whitespace.md |
| C 彩蛋长卷人生线 | 一条线串真实物件节点（竖版蜿蜒） | 个人经历、项目复盘、产品演化、成长路径 | styles/long-scroll.md |
选 B 时再追问一句用彩色档还是黑剪影档（剪影档≈高级冷感，详见该 style 文件）。 不想烧生图额度 / 要文字 100% 精确 → 走文末模式 B（HTML 文字卡）。
一、规格（固定）
- 3:4 竖版 1080×1440 px（A / B 出图、模式 B 渲染都按这个）。长卷 C 单图 3:4，节点多时出竖版长图 1080×N。
- ⚠️ 3:4 是硬要求。每次委派 Codex 生图，prompt 里必须反复强调"3:4 竖版 / 1080×1440"（订阅版 image_gen 不强调易给方图）；别要更高分辨率（易失败）。
- 第 1 张 = 封面（小红书首图）。
二、方法论底座（三风格共用，按需读取）
references/scene-patterns.md：6 类场景（拉扯/涌出/审查/重命名/筛选/加码）+ 原创隐喻三步 + 五步提炼 + 短标签规范 + shot-list 格式。references/master-lock.md：母版锁定——先锁锚点、抽不变量、写 ≥3 变异点；防元素清单化 / 母版复刻化；出图拦截规则。references/fact-anchoring.md：事实锚定——个人经历/品牌/数字只用作者给的，接本项目"私货清单"，反幻觉。references/qa-gates.md：两级质量门（CRITICAL/STANDARD）+ 逐张 QA + 落盘核验 + 生图稳定性经验。references/ip-character.md：IP 形象行为 DNA——形变弹性 / 禁止项 / 合格 vs 不合格动作（三风格共用）。references/prompt-scaffold.md：标准模式生图 prompt 骨架 + 负面约束块 + 三段局部编辑 prompt（长卷骨架在styles/long-scroll.md §六）。assets/masters/：作者自建视觉母版库（出图质量标尺）。首跑通常为空——正常，直接用 6 类场景抽象锚点，出过合格图再回填。
三、数量与切分
- 6-9 张：封面 1 + 内容若干 + 收尾 1。按"认知锚点"切，不要每段都配图。
- 认知锚点 = 核心判断 / 起因痛点 / 关键产品动作 / 前后对比 / 核心金句隐喻 / 边界判断 / 行动建议。
- 长卷 C 例外：整组就是一条人生线（5-8 节点），按
styles/long-scroll.md切节点。
四、标准流程
- 选风格（见〇）。加载对应
styles/*.md+ 方法论底座。 - 读料：Read
workspace/内容输出/<主题slug>/final.md（标题/金句/正文；若该主题只有final-formatted.md/final-candidate.md则读现有定稿版）+ 根目录workspace/voice-profile.md（文案守【绝对不要】、不"算账"字眼）。确认参考图workspace/个人知识库/个人IP图/个人IP动漫图.png在。用户没给主题 slug 就先问。 - 事实锚定：按
references/fact-anchoring.md过一遍——涉及亲历/品牌/数字的，缺口走私货清单或降级，别编。 - 写 shot-list（落
workspace/内容输出/<主题slug>/xiaohongshu-v2/shot-list.md）：每张先填母版锁定六字段（master-lock.md §三），再填用途/主题/共鸣点/物理动作/主物件/主角动作/短标签（scene-patterns.md §六格式）；出图 prompt 按references/prompt-scaffold.md（标准模式）或styles/long-scroll.md §六（长卷）。 4.5 shot-list 交主编确认（最便宜的纠偏点）：把切分 + 每张隐喻 / 物理动作 / 短标签交人类主编过一遍，认可方向后再出样张——对齐 shot-list 先行与本项目"初稿要人类确认"信条。 - 先出 2-3 张样张（封面 + 最核心隐喻 + 一张产品/对比页）委派
Agent（subagent_type: codex:codex-rescue），规范见〇五。 - 给人类主编验收样张（形象像不像、隐喻成不成立、中文有无错字、是否 3:4、风格对不对）。认可后再批量出剩余。
- QA：逐张过
references/qa-gates.md，不合格局部编辑或重生成。 - 配文产出：
caption.md（每张配的小红书正文段落）+titles.md（小红书标题候选，可调headline-craftskill）。供发布时贴正文与标题。 - 生图 prompt 先存后生（🔴 作者硬规矩，2026-06-11；时机在步骤 5 出样张之前）：调用生图引擎之前，先把每张完整生图 prompt 写进
xiaohongshu-v2/shot-list.md末尾「生图 prompt 存档」区块落盘（公共的通用前缀/后缀/负面块提取一次，每张只列差异化的画面+标题+批注），落完才允许出图；执行中临场改了 prompt，出图后把实际版本同步回档。这样换引擎/失败重试零成本、出错字直接改那条重出，prompt 绝不允许只活在 Bash 历史或委派指令里。
What ships with it
11 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- assets/masters/README.md 1.8 KB
- references/fact-anchoring.md 2.3 KB
- references/ip-character.md 4.2 KB
- references/master-lock.md 5.0 KB
- references/prompt-scaffold.md 5.6 KB
- references/qa-gates.md 5.7 KB
- references/scene-patterns.md 6.1 KB
- render_cards.mjs 7.0 KB runs code
- styles/long-scroll.md 5.0 KB
- styles/warm-anime.md 2.8 KB
- styles/whitespace.md 4.1 KB
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 109 lines · 124 tokens per session scan A 2bfbcdd8ad8d
article-to-xhs-cards is a skill published in the GitHub repository bailutingyu/OpenByline (2 stars, last pushed 2mo ago), licensed MIT. It adds 124 tokens to every session and 3,524 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
webgl-holographic-foil
A self-contained WebGL2 hero: thin-film interference over a crushed-foil surface whose palette shifts with the viewing angle; move the cursor to tilt the film.
general-video
Author or edit a custom HyperFrames composition when no specialized workflow fits, or when BRIEF.md sets flow: companion. Use for longer or multi-scene pieces, brand and sizzle reels, montages, static loops, static title cards, footage remixes, and freeform builds. Use motion-graphics instead for a short unnarrated…
html-ppt-hermes-cyber-terminal
OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.
html-ppt-taste-brutalist
16:9 HTML deck in tactical-telemetry / CRT-terminal taste. Deactivated-CRT charcoal slides, white-phosphor monospace, hazard-red accent, scanline overlay, ASCII syntax, density over decoration. Distilled from Leonxlnx/taste-skill brutalist-skill (Tactical Telemetry mode).
chengfeng-check-updates
An environment manager for a video-editing system. It checks whether its skills and runtime—the software needed to run them—are installed and compatible.
diagnostic-stem-delivery
Audio production with diagnostic analysis, timecode parsing from documents, and verified export workflow.