Token导航 LogoToken导航

Agent Skill · 图像处理

weixin-long-image(微信长图)

用于辅助图像生成、图片编辑、视觉素材处理或图像模型工作流。它适合让 Agent 根据文本生成图片、处理背景、整理视觉提示词或调用相关图像工具。使用时需要确认输入图片、版权来源、输出格式和模型限制;涉及人物、品牌、商品或公开展示素材时,应额外核对授权、真实性和内容合规边界。

weixin-long-image 是图像处理类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为操作浏览器;许可证为 MIT-0,第三方扫描有提醒项,按安装量在图像处理类的 1,562 个 Skill 里排第 648。

总安装 3,576周安装 149GitHub Stars 公开资料未说明下载 1,192核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
0.1.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:weixin-long-image(微信长图)
来源仓库:https://github.com/ssochi/weixin-long-image
安装命令:
openclaw skills install weixin-long-image
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install weixin-long-image

这个 Skill 是做什么的

图像处理类 Skill 处理生成、裁剪、压缩和批量转换图片的任务。

将微信聊天记录渲染成长 PNG 图像便于阅读分享。解决长文本超出屏幕显示范围的问题。weixin-long-image 属于图像处理类 Skill,可作为该场景下的辅助能力补充。自动分页拼接保持对话连贯性与视觉美观。输出为静态图片,不支持交互元素与动态更新。涉及他人消息时需获得同意以免侵犯隐私。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install weixin-long-image;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要操作浏览器,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

weixin-long-image 怎么安装?

在终端执行 openclaw skills install weixin-long-image。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

weixin-long-image 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

weixin-long-image 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

weixin-long-image 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

weixin-long-image 用的人多吗?

公开数据是累计安装 3,576 次,近一周安装 149 次,下载 1,192 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

weixin-long-image 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。

先想清楚的地方

  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理图像处理类任务、又不想每次都手写提示词的人。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

图像处理类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
ai-image-generation

AI 图像生成 · Codex、Claude

4,365,552382调用外部服务未明确有提醒
openai-image-gen

OpenAI 图像生成 · OpenClaw

1,906,88432需要联网MIT-0有提醒
image

图像 · OpenClaw

525,99926需要联网MIT-0通过
baoyu-cover-image

宝玉封面图 · Codex、Claude

474,24016,779只读MIT有提醒
baoyu-image-gen

宝玉图像生成器 · Codex、Claude

468,00716,778接触敏感数据MIT通过
weixin-long-image

微信长图 · 本页

3,576—操作浏览器MIT-0有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
weixin-long-image
description
>-

Weixin Long Image

Use this skill to turn rich HTML layouts into long PNGs for Weixin delivery.

Core Rule

In Weixin direct chats, if the reply body for the user would exceed 150 Chinese characters, treat this as a hard rule: prefer this skill over a long plain-text message.

When To Use

Use this skill when any of these is true:

  • The Weixin reply would be long and hard to read as plain chat text.
  • The user asks for a card, poster, beautiful layout, or visually polished output.
  • The content needs HTML rendering power, such as:

- charts - tables - timelines - curves / trend visuals - relation diagrams - dashboards - mixed text + image layouts

  • You want stable final presentation rather than chat-native formatting.

Default Visual Standard

Unless the user explicitly wants a poster/card/dashboard style, default to a WeChat public-article layout, not a centered floating card.

Night Rule

In Asia/Shanghai time, if the image is being prepared at 22:00 or later, default to a dark-mode article template to reduce eye strain, unless the user explicitly asks for a light theme.

Default article layout

  • Use a plain white page background.
  • Keep only moderate side margins for readability.
  • Do not wrap the whole article in a large rounded card with heavy shadow.
  • Prefer continuous reading flow, like a WeChat public account article.
  • Use readable typography, generous line height, and responsive images.

Default dark article layout after 22:00

  • Use a dark page background with softened contrast.
  • Keep the same article flow and margins as the daytime template.
  • Avoid excessive glow, neon accents, or dashboard-style visuals unless explicitly requested.
  • Preserve readability first: muted metadata, bright headings, comfortable code blocks, and responsive images.

Use card style only when appropriate

Reserve card/poster/dashboard styling for cases like:

  • cover cards
  • poster-like announcements
  • KPI / dashboard views
  • side-by-side comparison blocks
  • highly visual share cards

For ordinary long articles, analysis notes, translated writeups, and image-heavy explanations, prefer article layout.

Reusable Templates

Use assets/wechat-article-template.html as the default starting point for daytime article-style pages. Use assets/wechat-article-template-dark.html as the default starting point for article-style pages prepared at or after 22:00 Asia/Shanghai.

They provide:

  • WeChat-like article spacing
  • narrower side margins
  • no outer card shell
  • responsive images
  • readable tables
  • wrapped code blocks
  • simple note/tip blocks
  • matched light/dark article styles

Workflow

  1. Write a complete HTML document first.
  2. For article-like content, choose the template by time:

- before 22:00 Asia/Shanghai → assets/wechat-article-template.html - at or after 22:00 Asia/Shanghai → assets/wechat-article-template-dark.html

  1. Render the HTML with scripts/render_long_image.py.
  2. Send the PNG with the message tool using an absolute local path.
  3. After the message send succeeds, immediately delete temporary render artifacts (.png and temporary .html) unless the user explicitly asked to keep them.
  4. In Weixin direct chats, use message for any progress update; do not rely on multi-part assistant text arriving in order.

Quick Start

python3 scripts/render_long_image.py \
  --input /absolute/path/to/page.html \
  --png-out /absolute/path/to/output.png

You may also pass raw HTML via --input or stdin.

Input Rules

  • The script accepts complete HTML only.
  • --input can be a file path or inline HTML.
  • If --input is omitted, read HTML from stdin.
  • --html-out is optional.

- If provided, persist the rendered HTML there. - If omitted and --input is an existing HTML file, reuse that file. - If omitted and the source is inline HTML or stdin, create a temporary HTML file and auto-delete it after rendering.

  • If the source content starts as text, convert it into HTML first.

Rendering Guidance

  • Build for phone reading first.
  • For long-form articles, prefer article layout over card layout.
  • For tables, charts, diagrams, and mixed media, define layout explicitly in CSS instead of relying on browser defaults.
  • If images are embedded in the HTML, make them responsive with max-width: 100% unless overflow is intentional.
  • Use absolute output paths so the PNG can be delivered reliably.
  • If the content is sensitive, write outputs inside the workspace or /tmp, send the PNG, then remove temporary files if appropriate.

Sending

Use message with the generated PNG path.

  • action=send
  • channel=openclaw-weixin
  • media=/absolute/path/to/output.png
  • message=一句很短的说明

Post-send cleanup

For temporary outputs, once message reports success, delete the local render artifacts immediately.

Typical cleanup:

rm -f /absolute/path/to/output.png /absolute/path/to/output.html

Rules:

  • Prefer writing temporary outputs under /tmp.
  • Delete the PNG after a successful send unless the user asked to keep the file.
  • Delete the HTML too when it is only an intermediate artifact.
  • If render_long_image.py used inline HTML or stdin without --html-out, the script already auto-deletes its temporary HTML file; you still need to delete the PNG after send.

After a user-visible message send, reply with NO_REPLY to avoid duplicates.

Troubleshooting

  • If Playwright cannot launch Chromium, verify the local Playwright browser runtime is installed.
  • If the screenshot looks clipped, increase --width or fix the HTML/CSS layout first.
  • If the result is too tall, split the content into multiple HTML pages and render multiple images.
  • If the content is short and visually simple, skip this skill and reply normally.

Assets

  • assets/wechat-article-template.html: Default daytime HTML template for WeChat-article-style long images.
  • assets/wechat-article-template-dark.html: Default night-mode HTML template for WeChat-article-style long images after 22:00 Asia/Shanghai.

Script

  • scripts/render_long_image.py: Render complete HTML into a long PNG screenshot.

平台分布

OpenClaw

80.13%
按下载量换算955

安全审计

VirusTotal

通过

ClawScan

可疑

Static analysis

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills