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/wechat-cover-image/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/wechat-cover-image)<a href="https://agentmods.dev/skills/bailutingyu/openbyline/wechat-cover-image"><img src="https://agentmods.dev/badge/skills/bailutingyu/openbyline/wechat-cover-image.svg" alt="Measured on agentmods" 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 | $0.00094 | $0.02360 |
| Opus 5 | $0.00047 | $0.01180 |
| Sonnet 5 | $0.00019 | $0.00472 |
| Haiku 4.5 | $0.00009 | $0.00236 |
Grade A, and why
wechat-cover-image 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 8d 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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
公众号头图生成流程(个人 IP × 文章内容)
头图是公众号文章的"第一眼"。这套流程的信条:头图的灵魂是作者本人的 IP 形象——跨文章复用同一个 IP,形成辨识度;每篇头图 = 固定的 IP 形象 + 这篇文章的金句 + 主题场景。
一、头图规格(已核实,固定)
- 比例 2.35:1(头图硬要求),默认输出 1400×596 px(正好 2.35:1,公众号头图已足够清晰)。⚠️ 别追求更高分辨率(如 2350×1000)——订阅版生图分辨率太高容易直接生成失败;1400×596 是稳妥默认。(prompt 怎么写、出图后怎么量比例/裁切,操作步见 §四。)
- 微信裁剪安全区(关键):在订阅号列表/转发卡片里,封面会被自动裁掉两侧、只保留中间约 383×383 的正方形。所以——
- 核心元素(IP 形象 + 主标题大字)必须放在画面中央横向 ~60% 区域内,别贴左右边缘,否则缩略图里会被切掉。
- 左右两端只放背景/装饰,不放关键信息。
- 可选另备 1:1(500×500) 用于分享卡片。
二、前置:个人 IP 形象(没有就先建)
-
已有 IP 图 → 在
workspace/个人知识库/个人IP图/(作者个人视觉资产,跨主题复用、不进单篇主题目录——每篇文章复用同一套 IP)。生成头图时它是形象与配色的基准,每张头图里的人物都要和它一致(发型 / 眼镜 / 标志服饰 / brand 主色)。 -
还没有 IP 图 → 先引导作者用生图引擎(引擎选择守全局生图规则)生成一张 IP 设定图(character sheet),再挑一版满意的固定为资产。设定图生成 prompt 模板:
为一位【身份:如产品经理转型 AI 实战派】做一张个人 IP 角色设定图(character sheet),动漫手绘风、暖色调、亲和。包含:正面/侧面/背面三视图、6-8 个表情包(开心/思考/惊讶/认真/加油/专注…)、一排日常穿搭、一个主视图特写。人物特征:【年龄气质、发型、是否戴眼镜、标志性服饰与颜色、配饰】。brand 主色:【如橙色】。整体留白干净,像一页可复用的 IP 规范图。
让作者挑一版,之后所有头图都基于这版形象。
三、头图构图模板(从满意样例提炼)
横幅头图(规格见 §一),左右分区:
- 左 ~1/3:IP 形象——放在与文章主题强相关的场景里(如对着数据/敲电脑/看发布会),姿态呼应文章情绪。形象严格贴 IP 设定图。
- 中右 ~2/3:大字主标题——文章的核心金句/反差钩子,≤12 字/行、最多两行;关键词用 brand 主色高亮放大;下方可加一行小字副标题。
- 背景:与 IP brand 色统一的暗调 + 主色光效,可叠主题相关暗纹(科技电路 / 数据流 / 书房暖光)。
- 右上/右下角落(可选):一个呼应主题的小装饰元素(图标 / 清单卡片 / 对比小卡)。
- 文字要求:必须是清晰、可读的中文,别让模型把字画糊或画成乱码英文;字数宁少勿多。
- 🔴 标题文字逐字写进生图 prompt、让引擎一次性画进图里;禁止先出无字底图再用 PIL/脚本后期叠字(作者明确反感两步法)。文字调性活泼有力、关键词放大并用 brand 主色高亮、错落有节奏,别用呆板等线黑体/宋体。
文案提炼:主标题优先用文章里最有反差/最戳人的那句(金句或标题的浓缩),不要照搬整个长标题。
四、生成流程
引擎选择与操作守全局生图规则(用户级 CLAUDE.md 生图节)——弹窗选引擎、参考图缩图、Chrome 就绪确认等细节都在那边;所选引擎若最宽只到 16:9 → 先出 16:9,再走 step 5 居中裁(§三"核心元素垂直居中"让居中裁无损)。 🔴 每张头图必带个人 IP 参考图(
workspace/个人知识库/个人IP图/主形象图),人物 / 配色与之一致;IP 规范详见article-to-xhs-cards/references/ip-character.md§八。
- 读
final.md(取标题与金句)+个人知识库/个人IP图/(取形象与配色)。 - 提炼头图文案:主标题(金句/反差句,标出高亮词)+ 可选副标题。
- 组装生图 prompt = IP 形象详描 + §三构图模板 + 文案 + 主题场景 + brand 配色 + 中文文字清晰 + 尺寸/比例 + 构图约束(务必写明):
- 目标尺寸与比例直接写进 prompt:"输出 2.35:1 横幅 banner,尺寸 1400×596 px",并反复强调 2.35:1 比例(比尺寸更关键;规格依据见 §一);
- "把人物和所有文字安排在画面垂直居中的中间区域,上下各留出约六分之一高度的纯背景/光效延展,不要让人物头顶或文字贴近上下边缘"——这样万一工具退回 3:2 等固定画布、需要裁成 2.35:1 时也不会切到核心;
- "标题大字横向也别贴左右边,给微信缩略图的中间裁剪安全区留出空间"。
3.5 prompt 先存后生(🔴 作者硬规矩,2026-06-11):组装好的完整 prompt 先写进该文章目录的
头图-prompt存档.md落盘,落完才允许调用生图引擎;执行中临场改动,出图后把实际版本同步回档。换引擎/失败重试直接复用档内 prompt。
- 出图:按用户选定的引擎执行,调用方式 / 比例档位 / 禁 API 兜底等操作细节守全局生图规则(见本节开头);带上本节开头要求的 IP 参考图;生成后把图落到该文章目录
workspace/内容输出/<主题>/,文件名统一头图.png(多尺寸时如头图_2.35x1.png)。 - 取回后先量实际比例:已是 2.35:1(或足够接近)→ 直接缩放到
1400×596、无需裁;比例明显不对(工具退回1024×1024/1536×1024等)→ 再居中裁 2.35:1(高度 = 宽 ÷ 2.35;1536×1024→ 居中裁1536×654→ 缩放1400×596)。因为 step 3 已把核心预留在中间带,居中裁即无损。兜底裁切脚本(PIL,固定复用):
裁完 / 缩放后肉眼核对:人物脸 / 标题完整、中间 383 安全区信息齐、文字清晰、人物像 IP。from PIL import Image im = Image.open(src).convert("RGB"); w, h = im.size ch = round(w / 2.35); top = (h - ch) // 2 # 居中裁;核心若略偏上可手动减小 top im.crop((0, top, w, top + ch)).resize((1400, 596), Image.LANCZOS).save(out, quality=95) - 用真实图路径替换
final-formatted.md封面占位(真图即同目录头图.png,写成)。
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.
- 8d ago First seen · 77 lines · 94 tokens per session scan A 4fb763c00ed9
wechat-cover-image is a skill published in the GitHub repository bailutingyu/OpenByline (2 stars, last pushed 2mo ago), licensed MIT. It adds 94 tokens to every session and 2,360 once invoked, about $0.0005 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).
diagnostic-stem-delivery
Audio production with diagnostic analysis, timecode parsing from documents, and verified export workflow.
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.