Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add chengkj99/kj-skills --skill wechat-publish-kitgit clone --depth 1 https://github.com/chengkj99/kj-skillsWrote 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/chengkj99/kj-skills/wechat-publish-kit)<a href="https://agentmods.dev/skills/chengkj99/kj-skills/wechat-publish-kit"><img src="https://agentmods.dev/badge/skills/chengkj99/kj-skills/wechat-publish-kit/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/chengkj99/kj-skills/wechat-publish-kit"><img src="https://agentmods.dev/badge/skills/chengkj99/kj-skills/wechat-publish-kit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00141 | $0.03346 |
| Opus 5 | $0.00071 | $0.01673 |
| Sonnet 5 | $0.00028 | $0.00669 |
| Haiku 4.5 | $0.00014 | $0.00335 |
Grade A, and why
wechat-publish-kit 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 — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.
WeChat Publish Kit
把一篇文章从“正文已写好”推进到“可以发公众号”:生成配套文案、自动生成封面图、打包发布用 Markdown,引导用户从备选标题中选择一个标题,并在标题确认后继续把标题、正文、封面图片、摘要简介保存为公众号草稿。
Direct-publish shortcut
用户已经给出最终标题,或直接说“发布到公众号草稿箱 / 保存草稿 / 用 API 发布”时,不再要求他重新选择标题,直接进入发布流程;但不能跳过封面生成、摘要检查和 API dry run。
直接发布仍按以下顺序执行:
- 检查最终标题、正文与摘要;
- 解析封面路径并执行封面预检;
- 没有合格封面时自动生成;
- 运行 API dry run;
- 创建新草稿,或使用已知
media_id更新原草稿。
Core Workflow
-
Read the source article
- Prefer the canonical source the user points to:
.md,.docx, or an existing article draft. - If both Word and Markdown exist, use Markdown for automated publishing and treat Word as the human-facing version unless the user says otherwise.
- Extract the article positioning, target reader, pain point, concrete promise, and any existing title/frontmatter.
- If the source already has a strong title, keep it as one of the title options instead of silently replacing it.
- Prefer the canonical source the user points to:
-
Generate companion materials
- Article intro: 150-300 Chinese characters, reader-pain first, no “本文将介绍”.
- 5 title options unless the user has already chosen one.
- Summary: under 120 Chinese characters for WeChat digest / social preview.
- Cover prompt: English, explicit 2.35:1 /
--ar 47:20, with subject-specific visual elements. - Cover image path:
raw/studio/funnel/formatted/wechat/assets/<yyyy-mm-dd-slug>/cover.png. - Forward copy: one Moments version and one AI practice group version.
- Companion file must include:
- source article path
- publish copy path
- chosen-title status:
待选择until user confirms - generated cover image path
- cover prompt
- title options
- summary
- article intro
- forward copy
-
Generate the cover image by default
- Default behavior: generate a usable cover image, not only a prompt。即使用户直接要求发布草稿、没有经过标题选择环节,也必须执行本步骤。
- Resolve the cover in this order:
- 用户明确指定的封面;
- 当前文章资产目录里已经存在且尺寸合格的
cover.png; - 根据最终标题、文章核心冲突和读者对象自动生成的新封面。
- 正文首图、数据图或流程图不自动等同于封面。只有用户明确指定,或它本身就是已验证的 2.35:1 封面时才能复用,避免发布脚本静默采用正文第一张图。
- Use the image generation tool after drafting the cover prompt.
- Store the final usable cover under the article’s WeChat asset folder:
raw/studio/funnel/formatted/wechat/assets/<yyyy-mm-dd-slug>/cover.png. - If the generated image is returned outside the workspace, copy or move the final selected image into the asset folder.
- Verify the file exists and inspect image metadata with
file,sips -g pixelWidth -g pixelHeight, or an equivalent metadata command. - Generate a wide composition close to 2.35:1, then normalize the final file to
900 × 383when local image tooling is available. Keep the subject and visual conflict inside the safe center area so desktop and mobile crops both work. - 默认不把文章长标题画进图片,避免错字和移动端缩小后不可读;公众号标题栏会承载文字。只有用户明确要求“封面带字”时才生成文字版。
- 封面要表达文章的核心冲突或结果,不要只堆 AI Logo、机器人、代码雨或通用科技元素。
- Only skip image generation when the user explicitly says “只要提示词 / 不生成图片 / no image”.
- If image generation is unavailable, keep the cover prompt and mark the companion file
cover_status: 未生成(缺少可用图像生成工具); do not pretend a cover exists. - Before publishing, verify:
cover.pngexists;- dimensions are
900 × 383or aspect ratio is close to 2.35:1; - frontmatter
coverresolves to this file; - API receives this file as its thumb/cover argument.
What ships with it
3 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.
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 · 203 lines · 141 tokens per session scan A 31dbf517fa73
wechat-publish-kit is a skill published in the GitHub repository chengkj99/kj-skills (14 stars, last pushed 11d ago), licensed MIT. It adds 141 tokens to every session and 3,346 once invoked, about $0.0007 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…