← 返回 Skills 市场
112
总下载
0
收藏
1
当前安装
4
版本数
在 OpenClaw 中安装
/install wechat-oa
功能描述
微信公众号草稿箱管理工具集。触发词（满足任一即触发）：看看草稿箱/查看草稿/草稿列表/公众号草稿/搜草稿/搜索草稿/按关键词找草稿/按标题搜/创建草稿/新建草稿/发文章到公众号/推送文章/更新草稿/删除草稿/批量删除草稿/生成封面图/上传图片到公众号/上传图片到素材库/已发布文章列表/公众号素材列表/素材管理/删...
安全使用建议
This skill appears to implement what it says (WeChat draft/material management and image generation) but contains metadata/instruction mismatches you should address before use:
- Verify the source: the registry 'Source' is unknown while SKILL.md/README reference a GitHub repo. Confirm the code origin and trustworthiness before running.
- Credentials: The tool requires your WeChat AppID and AppSecret stored in config.json. Do NOT commit that file to any remote repo. Only provide these credentials in a secure environment and consider using least-privilege accounts.
- Dependencies: Manually install required Python packages (requests, Pillow; premailer is optional) in a virtual environment. The registry did not declare these — the skill will fail otherwise.
- IP whitelist: If your WeChat account uses IP whitelisting, you'll need to add the machine's public IP (instructions reference curl ifconfig.me). Be aware this queries an external service to get your public IP.
- Safety: Review the two included Python files yourself (or have a trusted reviewer do so). Network calls in the code target official WeChat endpoints; there are no obvious hidden remote endpoints, but the metadata inconsistencies lower trust.
If you plan to use this skill in an autonomous agent, remember the agent will be able to run the code and thus access any local config.json containing credentials — only enable autonomous invocation if you trust the skill and its origin.
功能分析
Type: OpenClaw Skill
Name: wechat-oa
Version: 1.5.0
The wechat-oa skill bundle is a legitimate toolkit for managing WeChat Official Account drafts and materials. The core logic in wechat_push.py and generate_infographic.py interacts exclusively with official WeChat API endpoints (api.weixin.qq.com) to perform administrative tasks like draft creation, image uploads, and subscriber management. The code uses standard libraries (requests, Pillow) and follows the stated purpose without any evidence of data exfiltration, malicious execution, or harmful prompt injection.
能力标签
能力评估
Purpose & Capability
The code (wechat_push.py + generate_infographic.py) implements WeChat draft/material management and image generation which fits the stated purpose. However the SKILL.md and README claim "no third‑party dependencies" while the code clearly uses external Python packages (requests, Pillow, optional premailer). The registry metadata lists no required env vars or primary credential, but the tool requires AppID/AppSecret stored in config.json. These mismatches between claims and actual needs are concerning.
Instruction Scope
Runtime instructions require copying/editing config.json (containing AppID/AppSecret), adding your server IP to WeChat's IP whitelist, and reading design.md before generating HTML. The instructions also suggest querying an external endpoint (curl ifconfig.me) to find your public IP. Other than that, the code's network calls appear limited to official WeChat API endpoints. The instructions do require reading local files (design.md, config.json) and writing generated images to disk, which is expected for this tool.
Install Mechanism
No install spec is provided (instruction-only in registry), so the skill will not install automatically. However the README instructs installing Python packages (requests, Pillow) and the code imports premailer optionally — these dependencies are not declared in the registry metadata. Because there is no automated install step, users must manually install packages; ensure you install them from trusted package indexes and in an isolated environment.
Credentials
The tool legitimately requires WeChat credentials (AppID and AppSecret) to call official APIs, and it stores them in a local config.json. But the skill metadata does not declare these credentials or mark a primary credential, which is inconsistent and can lead to accidental credential exposure. The number and nature of credentials requested (AppID/AppSecret) are proportional to the purpose, but the omission from the declared requirements is a red flag.
Persistence & Privilege
The skill does not request always:true and is user-invocable only. It reads/writes local files (config.json, generated images) and does network calls to WeChat APIs; that level of persistence and privilege is appropriate for a client-side WeChat management tool.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install wechat-oa - 安装完成后，直接呼叫该 Skill 的名称或使用
/wechat-oa触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.5.0
Auto update from GitHub
v1.4.0
Auto update from GitHub
v1.3.1
Auto update from GitHub
v1.3.0
wechat-oa 1.3.0
- 增加详细的命令及参数说明，涵盖草稿和素材的批量管理、自动化图片处理等功能。
- 丰富依赖与初始化配置步骤，包含 IP 白名单、凭证配置、安全注意事项。
- 明确封面图生成、正文图片自动上传、智能摘要等核心特性。
- 增补官方 API 映射表，详细列出每个命令的对应接口。
- 新增排版与内容规范，特别对 HTML/CSS 兼容性作详细说明。
- 完善使用示例与输出文件说明，便于新用户快速上手。
| 命令 Command | 说明 Description | 底层API Underlying API |
|------|------|---------|
| `list` | 查看草稿列表（含标题+时间）/ View draft list with title+time | `draft/batchget` |
| `find <关键词>` | 按标题关键词搜索草稿 / Search drafts by title keyword | `draft/batchget` |
| `create <文件>` | 创建新草稿（支持 .html 和 .md，自动上传正文配图）/ Create draft from HTML or MD (auto-upload inline images) | `draft/add` + 永久素材上传 |
| `update <media_id> <文件>` | 更新已有草稿（自动上传正文配图）/ Update existing draft (auto-upload inline images) | `draft/update` |
| `update <media_id> <文件> --force-cover` | 更新草稿并强制重新生成封面 / Update draft + force-regenerate cover | `draft/update` |
| `delete <media_id>` | 删除草稿 / Delete draft | `draft/delete` |
| `batch-del <id1> [id2] ...` | 批量删除草稿 / Batch delete drafts | `draft/delete` |
| `upload <图片文件>` | 上传图片到永久素材库 / Upload image to permanent material | `material/add_material` |
| `materialcount` | 获取各类永久素材总数 / Get permanent material count by type | `material/get_materialcount` |
| `materials [type] [count] [offset] [keyword]` | 批量获取素材列表，支持关键词过滤 / Batch list with keyword filter | `material/batchget_material` |
| `materialdel [media_id...]` | 批量删除素材（多个空格分隔）/ Batch delete materials | `material/del_material` |
| `materialdel` | 交互式删除（列出→选编号→确认）/ Interactive material deletion | `material/del_material` |
| `materialdel <type>` | 交互式删除指定类型 / Interactive deletion by type (news/image/video/voice) | `material/del_material` |
| `published` | 获取已发布文章列表 / List published articles | `material/batchget_material` |
| `cover <标题>` | 生成封面图预览（不推送）/ Generate cover preview (no push) | PIL local generation |
| `userinfo <openid>` | 获取用户基本信息（需认证账号）/ Get user info by openid | `user/info` |
| `userlist [next_openid]` | 获取用户列表（需认证账号）/ List all users | `user/get` |
元数据
常见问题
Wechat Oa 是什么？
微信公众号草稿箱管理工具集。触发词（满足任一即触发）：看看草稿箱/查看草稿/草稿列表/公众号草稿/搜草稿/搜索草稿/按关键词找草稿/按标题搜/创建草稿/新建草稿/发文章到公众号/推送文章/更新草稿/删除草稿/批量删除草稿/生成封面图/上传图片到公众号/上传图片到素材库/已发布文章列表/公众号素材列表/素材管理/删... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 112 次。
如何安装 Wechat Oa？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install wechat-oa」即可一键安装，无需额外配置。
Wechat Oa 是免费的吗？
是的，Wechat Oa 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Wechat Oa 支持哪些平台？
Wechat Oa 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Wechat Oa？
由 Woody（@andy8663）开发并维护，当前版本 v1.5.0。
推荐 Skills