yb2460

CLI harness for WPS Office -- let AI agents control Writer, Calc & Impress via COM automation

15
0
89% credibility
Found May 27, 2026 at 15 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

This is a command-line tool that lets you control WPS Office (the popular Chinese office suite) through simple text commands. It provides 47 commands for creating, editing, and exporting documents, spreadsheets, and presentations. The tool is designed so AI assistants can automate office tasks just as easily as they can work with open-source software. You can use it interactively from the terminal or hand it to an AI agent for batch automation. Everything runs through the real WPS Office engine, so the output looks exactly like documents created manually.

How It Works

1
💡 You need to automate WPS Office

You discover this tool when looking for a way to control WPS Office programmatically, perhaps for batch report generation or AI agent integration.

2
🛠️ You install the tool

You install it with a simple command, and it works alongside your already-installed WPS Office on Windows.

3
📄 You create a new document project

You start a new project by telling the tool what kind of document you want: a text document, spreadsheet, or presentation.

4
✍️ You add and edit content

You add paragraphs, headings, tables, and images by typing simple commands. Everything you type gets added to your document automatically.

5
🤖 Your AI assistant takes over

Here's the exciting part: you hand this tool to an AI assistant, and it can now create and edit WPS documents on your behalf, exactly like it would with open-source software.

6
You choose how to export
📕
Text documents

Export to DOCX, PDF, TXT, HTML, or RTF formats

📊
Spreadsheets

Export to XLSX, CSV, PDF, or HTML formats

📽️
Presentations

Export to PPTX or PDF formats

Your document is ready

Your document appears exactly as you planned, rendered by the real WPS Office engine with perfect formatting, ready to share or print.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 15 to 15 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is cli-anything-wps?

cli-anything-wps is a Python CLI tool that lets AI agents control WPS Office through 47 terminal commands. It wraps the WPS COM automation interface, enabling programmatic creation and editing of documents, spreadsheets, and presentations without touching the GUI. You run commands like `cli-anything-wps writer add-paragraph --text "Hello"` to manipulate real WPS files, and everything supports JSON output for agent parsing. The tool runs entirely on Windows via COM automation, matching the MS Office VBA interface.

Why is it gaining traction?

The hook is extending the AI-controllable software ecosystem. CLI-Anything already handles GIMP, Blender, and LibreOffice for agents - but those are all open-source. This project proves the same harness methodology works on closed-source software like WPS, the dominant office suite in China. Key features users notice: session state persistence with undo/redo (50 steps), `--json` flag for structured output, interactive REPL mode, and background execution for batch automation. The export pipeline converts projects to DOCX, XLSX, PPTX, PDF, and more using actual WPS rendering rather than file format hacks.

Who should use this?

AI agent developers building office automation workflows will find this most valuable. DevOps teams needing to generate reports or convert document formats in pipelines can use the background mode for silent batch processing. QA engineers testing WPS document processing can script reproducible test scenarios. Businesses locked into WPS Office that need programmatic document manipulation without licensing MS Office.

Verdict

At 15 stars, this is early-stage but the implementation is thoughtful - solid architecture with session management, clear documentation, and proper COM error handling. The credibility score of approximately 0.9% reflects its newness, so production use requires caution. Good fit for internal tools and agent experimentation. Watch for maturity before depending on it for critical workflows.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.