一键生成PPT截图和缩略图工具发布到微信去的工具，MAC版本

自动将指定Mac路径下的PPT按页生成截图和缩略图，并发布到微信公众号，支持自定义配置和参数调整。

Security Scan

The skill does what it claims: converts a PPT to images/thumbnails and publishes to the official WeChat APIs using credentials supplied in config.json; no unrelated credentials, endpoints, or hidden behavior were found.

Purpose & Capability

Name/description match the implementation: code converts PPT→PDF via LibreOffice, PDF→PNG via Ghostscript, builds thumbnails, and uploads/publishes to WeChat API endpoints. Required binaries (LibreOffice, Ghostscript) and Python libs (Pillow, requests) are appropriate for the task.

Instruction Scope

SKILL.md instructs the user to set ppt_dir/ppt_file and WeChat credentials in config.json and to install ghostscript and pillow. The runtime code only reads config.json, the specified PPT file, and calls the official WeChat API endpoints (api.weixin.qq.com); it does not access unrelated system files or unusual network endpoints.

Install Mechanism

There is no automated installer. SKILL.md recommends brew install ghostscript and pip3 install pillow — standard, low-risk instructions. No downloads from unknown hosts or archive extraction are present.

Credentials

The skill does not request environment variables but expects WeChat appid/secret in config.json. That is functionally appropriate, but storing secrets in a plaintext config file is a security consideration: do not commit config.json to source control and protect access to the file.

Persistence & Privilege

Skill is not always-enabled and does not modify other skills or system-wide configuration. It runs on demand and only uses local files and network calls to the WeChat API.

Assessment

This skill appears internally consistent and implements the stated workflow. Before installing/using it: (1) Verify you have LibreOffice (soffice) and Ghostscript installed and configured paths in config.json. (2) Replace the placeholder wechat_appid and wechat_appsecret in config.json with your real credentials and keep that file private (do not commit it to git). (3) Understand the script will upload images to your WeChat account using those credentials—only run it with credentials you control. (4) The script compresses images in-place and writes output under the PPT-named work directory; verify disk/write permissions and available space. (5) If you need higher assurance, review or run the script in an isolated environment (local VM) and monitor outgoing network calls; the only external endpoints used are api.weixin.qq.com (official WeChat API).

--- name: ppt2wechat description: mac系统某路径下某名字的PPT自动按页生成每一张截图和缩略图，然后发布到微信公众号上 ---

1、提示词输入：

例如，使用 ppt2png skill , MAC系统，桌面路径下10只小青蛙迎接春天.pptx，帮我生成图片和3行3列的缩略图，然后发布到微信公众号。我的微信公众号密钥：wechat_appid=XXX，wechat_appsecret=XXX

2、config.json配置说明：

路径赋值给config.json 下的 ppt_dir

名字赋值给ppt_file

需要用户说明自己的wechat_appid，wechat_appsecret

微信公众号如需更改作者也可以更改：wechat_author

3、skill实现说明：

基于PPT → PDF（使用 LibreOffice） PDF → 每页 PNG（使用 Ghostscript） 每 9张图片生成 1 张缩略拼接图（3列 × 3行），可通过调整thumbnail_cols列，thumbnail_rows行来实现 兼容 Python 3.6

4、需要提前安装： brew install ghostscript pip3 install pillow

5、工作流程

主程序：ppt2wechat.py

配置文件：config.json

自动读取配置

支持修改 DPI、列数、行数、LibreOffice 路径、gs 路径

针对config.json 如果 which gs 输出是： /opt/homebrew/bin/gs

就改成：

"ghostscript_path": "/opt/homebrew/bin/gs"