export-csv

name	export-csv
description	Export a CSV inventory of all documentation pages (title, navigation path levels, and public URL) from the frontmatter of every Markdown file under src/content
user-invocable	true

Export a CSV inventory of all documentation pages

Generate a flat CSV listing every documentation page, so the content can be reviewed, filtered, or analyzed in a spreadsheet.

A helper script does the work in code — don't assemble the inventory by hand.

What it produces

A CSV file (export.csv by default) with one row per page and these columns:

• title — from the page's frontmatter
• diataxis_content_type — the page's Diátaxis type (tutorial, how-to-guide, reference, explanation, or none); empty for list pages and untagged pages
• path_l1 through path_l6 — the first six URL path segments, i.e. the navigation hierarchy
• url — the public docs URL (https://docs.giantswarm.io/<relpath>)

The script walks all .md files under src/content, reads each file's YAML frontmatter, and derives the path levels and URL from the file location.

By default it excludes the /changes/ section (changelogs, path_l1=changes), which otherwise dominates the output. Adjust EXCLUDED_SECTIONS in the script to change which top-level sections are skipped.

Steps

1. Run the script (pure Python plus PyYAML, no Docker needed):

   python3 .claude/skills/export-csv/export_csv.py
   

   It writes export.csv to the current working directory and covers all pages under src/content.

2. Report the result to the user — the output path and how many rows were written. Offer to open or summarize the CSV if useful.

Notes

• The same script also backs the make export-csv target, which runs it inside the docs-scriptrunner Docker image (no local Python required).
• Pages with empty content or missing frontmatter are skipped.