A Claude Code skill that audits a Substack export ZIP before you migrate to another newsletter platform. Companion to the Substack Migration Checklist.
Point it at the ZIP you downloaded from Substack and it produces, fully offline:
report.md— export integrity, subscriber list audit, and post archive dependencies, with exact counts, plus the manual steps that remain.subscribers-cleaned.csv— import-ready list (malformed and duplicate emails removed, disabled addresses excluded, paid status flagged), matching the checklist's platform-agnostic template.subscribers-excluded.csv— every excluded contact and why, for review.posts-inventory.csv— per post: Substack-hosted images, substack.com links, embeds, paywall status — everything that needs attention when moving your archive.
Your original export is never modified.
Everything runs locally. Subscriber data is processed only by a
dependency-free Python script on your machine; the AI model only sees
aggregate counts and redacted samples (a***@example.com). No network access.
- Claude Code
- Python 3.9+ (standard library only — nothing to
pip install)
As a plugin (recommended — gets updates through the plugin system):
/plugin marketplace add david-is-back/substack-migration-skill
/plugin install substack-migration-check@substack-migration-check
Or as a personal skill:
git clone https://github.com/david-is-back/substack-migration-skill ~/.claude/skills/substack-migration-checkThe target directory name must be substack-migration-check (the skill name).
In Claude Code:
Check my Substack export at ~/Downloads/my-export.zip — I'm migrating.
Artifacts land in migration-check/ next to your ZIP.
python -m unittest discover -s tests -vMIT licensed. Issues and PRs welcome.