This repository packages a Codex skill for Python-driven OriginPro plotting. It helps Codex create reproducible originpro scripts, build Origin workbooks and graph pages, export publication figures, and save .opju projects from Excel or CSV data.

python-origin-plotting-skill/
  README.md
  AGENTS.md
  skills/
    python-origin-plotting/
      SKILL.md
      .clawhubignore
      agents/openai.yaml
      scripts/originpro_plot_template.py
      references/originpro-workflow.md
      references/troubleshooting.md

Copy the skill folder into your Codex skills directory.

PowerShell:

$repo = (Get-Location).Path
$dest = "$env:USERPROFILE\.codex\skills\python-origin-plotting"
New-Item -ItemType Directory -Force -Path (Split-Path $dest)
Copy-Item -Recurse -Force "$repo\skills\python-origin-plotting" $dest

macOS/Linux:

repo="/path/to/python-origin-plotting-skill"
dest="${CODEX_HOME:-$HOME/.codex}/skills/python-origin-plotting"
mkdir -p "$(dirname "$dest")"
cp -R "$repo/skills/python-origin-plotting" "$dest"

Restart Codex after installing, then try:

Use $python-origin-plotting to make an OriginPro grouped bar chart from this Excel file and save PNG plus OPJU outputs.

OpenClaw/ClawHub also uses a SKILL.md-based skill folder. This repository keeps the skill at:

skills/python-origin-plotting

To publish the skill bundle to ClawHub, publish that skill folder, not the whole repository:

npx clawhub whoami
npx clawhub publish .\skills\python-origin-plotting `
  --slug python-origin-plotting `
  --name "Python Origin Plotting" `
  --version 0.1.0 `
  --changelog "Initial release" `
  --tags latest

After publishing, inspect it:

npx clawhub inspect python-origin-plotting
npx clawhub inspect python-origin-plotting --files

For generic agents that do not understand Codex skills directly, point them to AGENTS.md or say:

Read AGENTS.md, then use the python-origin-plotting skill to generate an OriginPro plotting script from my Excel/CSV data.

The bundled script is a starting point for new OriginPro automation:

python .\skills\python-origin-plotting\scripts\originpro_plot_template.py .\data.xlsx --sheet "Sheet1" --out .\plots_origin --show

For real work, adapt prepare_data() and build_origin_project() to the target columns and graph types.

From this repository root:

python D:\.codex\skills\.system\skill-creator\scripts\quick_validate.py .\skills\python-origin-plotting
python -m py_compile .\skills\python-origin-plotting\scripts\originpro_plot_template.py

The first command checks skill metadata and structure. The second catches Python syntax issues without needing Origin to be installed.

1. Create a new empty repository on GitHub, for example python-origin-plotting-skill. Do not add a README on GitHub if this local folder already has one.
2. Open PowerShell in this folder:

cd "path\to\python-origin-plotting-skill"
git init
git add .
git commit -m "Add python-origin-plotting Codex skill"
git branch -M main
git remote add origin https://github.com/<your-name>/python-origin-plotting-skill.git
git push -u origin main

3. Replace <your-name> with your GitHub username. If Git asks you to sign in, follow the browser prompt or use a personal access token.

• Keep private data files, unpublished figures, and generated .opju outputs out of the repo.
• Commit only the reusable skill, template script, and documentation.
• Add screenshots to the GitHub README only if they are safe to publish.
• Tag a first version after the initial push:

git tag v0.1.0
git push origin v0.1.0