github-sanitize

name

github-sanitize

description

Sanitize n8n workflows and documentation before publishing to GitHub. Use this skill whenever the user mentions publishing to GitHub, pushing to public repo, sanitizing workflows, removing sensitive data, cleaning credentials, preparing files for GitHub, push to GitHub, publish workflows, ลบข้อมูล sensitive, push GitHub, publish, sanitize, ก่อน push, ก่อน publish, ทำความสะอาดข้อมูล, ลบ credential, or any action involving making n8n workflow files safe for public visibility.

GitHub Publish Sanitization

Use this skill before pushing n8n workflows to public GitHub to remove sensitive data.

What This Does

This skill automates the sanitization process using the existing sanitize.py script:

1. Dry-run check - Shows what will be changed without modifying files
2. User confirmation - Asks before applying changes
3. Sanitization - Replaces sensitive data with placeholders
4. Verification - Confirms no sensitive data remains
5. Next steps - Guidance for git commit & push

Prerequisites

Ensure .env.sanitizer exists and is configured with your real sensitive data:

cp .env.sanitizer.example .env.sanitizer
# Edit .env.sanitizer with your real values

Step-by-Step Execution

1. Dry Run (Preview Changes)

python sanitize.py --dry-run

This shows what will be changed. Review the output carefully.

2. Confirm & Sanitize

If dry-run looks good, run the actual sanitization:

python sanitize.py

The script will:

• Clear pinData fields in workflow JSON files
• Replace database names, URLs, emails, company names
• Update markdown documentation
• Verify results

3. Verify No Sensitive Data Remains

The script automatically runs verification. You can also run:

python sanitize.py --verify-only

4. Check Git Status

git status

Review which files were modified.

5. Commit & Push

git add .
git commit -m "docs: update workflows and sanitization"
git push origin main

What Gets Sanitized

Based on your .env.sanitizer configuration:

• Database names → YourDatabase
• Table names → YourTable
• URLs → example.com
• Emails → [e-mail]
• Company names → [Company Name] or YourName
• Credential IDs → YOUR_CREDENTIAL_ID

What's Preserved

• Node names - n8n workflow structure
• Node types - workflow functionality
• Connection IDs - workflow node links
• Instance IDs - but replaced with YOUR_INSTANCE_ID
• Empty pinData - {} objects remain

Files Processed

• workflows/*.json - n8n workflow files
• *.md - documentation files (README, AGENTS, etc.)

Troubleshooting

Q: Dry-run shows no changes but I know there's sensitive data? A: Ensure .env.sanitizer has the SANITIZE_ entries for your real data.

Q: Verification still finds sensitive data? A: Add new entries to .env.sanitizer with SANITIZE_ and PLACEHOLDER_ pairs.

Q: Pin data keeps reappearing after n8n export? A: Clear pinData manually in n8n editor before exporting, or run this script again.

Reference Documentation

For detailed step-by-step guide, see: docs/GITHUB_PUBLISH_GUIDE.md