automation

harness-anything

Name: harness-anything
Author: yb2460

Harness Anything - AI agent control hub: WPS, MS Office, Zotero, Photoshop, 47 CLI commands, 27 academic skills, SVG-to-PPTX

2,420 stars+113★ this week172 forksPythonUpdated 10/9/2026100% free · open source

What it does

Harness Anything lets you run AI‑driven automation across WPS, MS Office, Zotero, Photoshop, CLI tools and academic workflows from a single Python‑based hub.

When to use it

•You need to generate a PowerPoint deck from a batch of SVG diagrams without manually copying each image.
•You want to bulk‑export Zotero bibliography entries into a formatted Word report.
•You have repetitive image‑editing tasks in Photoshop that can be described in natural language and executed automatically.

Ready-to-paste prompt

harness run svg_to_pptx --input ./research/figures/ --output ./deck/StudyResults.pptx --title "2024 Clinical Trial Findings" --notes "Add speaker notes summarizing each figure"

Heads up: The tool requires a valid OpenAI API key **and** the target desktop applications (e.g., Microsoft Office, Photoshop) to be installed and accessible; without them the corresponding skills will fail at runtime.

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🖥️ Needs desktop apps

Drives software that must already be installed on your computer.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/harness-anything && curl -fsSL https://workflowstacks.com/api/skills/harness-anything/claude-skill -o ~/.claude/skills/harness-anything/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How harness-anything works

Codeflow

Free to inspect

Harness-anything is a large Python project (~16k lines across 93 code files, plus 11 test files). Setup is light: installs like a normal app. Reading the code is optional. Last commit a month ago, MIT license, has a test suite.

Size

Large codebase

~16k lines · 93 code files · ~2 h to skim

Setup

One-command install

Installs like a normal app. Reading the code is optional.

Runs on

Python

No API keys detected

Python 100%

Where to start reading

1
README.md
Start here — what it does and how to install it

What's in each folder

WPS/Folder71 files

cli_anything/Folder50 files

photoshop-harness/Folder19 files

illustrator-harness/Folder16 files

READMEHas testsDocumentedNo CIMIT licenseUpdated this month