Email List Cleaning Skill

Clean an email list on your own machine before you send or import it. Every removal gets a reason. Works as a Claude Code plugin (skill email-list-cleaning) or as a plain command-line tool. Python 3.9+, no dependencies, no account, and the list never leaves your computer.

python3 -m emailclean clean subscribers.csv --out cleaned/ --suppress unsubscribes.csv --suppress bounces.csv
./demo.sh        # runs the bundled example
./run_tests.sh   # 15 tests

In Claude Code: install the plugin, then ask "clean subscribers.csv and drop everyone in unsubscribes.csv".

Output: clean.csv (your original columns, same delimiter), review.csv and removed.csv (plus clean_status, clean_flags, clean_note), and summary.json.

Reads comma, semicolon, tab or pipe CSV, with or without a BOM. Finds the email column automatically ("Email Address", "E-mail", "email"...), or use --column. It also strips mailto: and <>, and keeps international domains like münchen.de as written (checks use punycode).

• python3 -m emailclean check a@b.com info@x.io: checks a few addresses.
• python3 -m emailclean update-disposable: refreshes the disposable list from the upstream CC0 project (disposable-email-domains (https://github.com/disposable-email-domains/disposable-email-domains)). --from file.txt uses your own list instead. It refuses lists under 1,000 domains.
• --disposable-list mine.txt: uses your own list for one run.

• It doesn't send email and doesn't check mailboxes over SMTP, so it can't prove an address exists. It removes what can be found without contacting anyone. A clean result is not a deliverability guarantee.
• The typo list only covers common consumer providers. Unknown business domains are never "corrected".
• Duplicates match local parts case-insensitively, which is how real providers behave even though the standard allows case-sensitive mailboxes.
• Follow the consent and unsubscribe rules that apply to you. This tool only helps you honor suppression lists.