Evernote to Markdown Migration Toolkit

Migrate thousands of Evernote notes to organized Markdown files in minutes, not hours.

Built by Evangel Oputa at Begine Fusion & OnStack AI Labs after migrating 6,453 notes from 293 Evernote notebooks in a single session.

---

The Problem

Evernote makes it nearly impossible to leave:

• No bulk export (manual, one-notebook-at-a-time clicking)
• No MCP server for AI integration
• Deprecated developer tokens
• Proprietary format (HTML-in-XML)
• CA$325/year for something your existing cloud storage can do

If you have hundreds of notebooks and thousands of notes, manual migration would take 15-20 hours of clicking. This toolkit does it in under 30 minutes.

What This Toolkit Does

Evernote Account
      |
      v
[1. Bulk Download] --> SQLite DB (all notes via API)
      |
      v
[2. ENEX Export] --> .enex files (one per notebook)
      |
      v
[3. Convert to Markdown] --> Organized .md files with YAML frontmatter
      |
      v
[4. Set Up Vault] --> Obsidian-ready system with search, templates, AI export
      |
      v
[5. Organize Stacks] --> Evernote-style Stack > Notebook > Note hierarchy

What You Get

• Every note as a clean .md file with YAML metadata (title, created, updated, tags, notebook)
• Every attachment extracted and linked (images, PDFs, files)
• Notebook structure preserved as folders
• Stack organization (group folders into stacks, just like Evernote)
• Full-text search tool (replaces Evernote search)
• Quick note capture tool (replaces Evernote quick note)
• AI export tool (JSON/JSONL for embeddings, RAG, LLM pipelines)
• Obsidian configuration (templates, bookmarks, daily notes)
• Works with any Markdown editor (Obsidian, VS Code, Typora, etc.)

Requirements

• Python 3.10+ (tested on 3.11)
• Evernote account (any plan)
• ~10 minutes for download + conversion

Quick Start

1. Clone and Install

git clone https://github.com/beginefusion/evernote-to-markdown.git
cd evernote-to-markdown
pip install -r requirements.txt

2. Get Your Evernote Auth Token

Option A: Developer Token (if available)

1. Go to https://www.evernote.com/api/DeveloperToken.action
2. Sign in and create a token
3. Copy the token string

Option B: Browser Cookie (works for all accounts)

1. Open https://www.evernote.com/client/ in your browser
2. Sign in to your account
3. Open DevTools (F12) > Application tab > Cookies
4. Find the cookie with a value starting with S=s###:U=...
5. Copy that value

3. Run the Migration

# Step 1: Download all notes from Evernote
python scripts/export_evernote.py

# Step 2: Convert to Markdown
python scripts/convert_to_markdown.py --input ./enex_export --output ~/NoteVault

# Step 3: Set up your note system
python scripts/setup_vault.py --vault ~/NoteVault

# Step 4 (optional): Organize into stacks
python scripts/organize_stacks.py --vault ~/NoteVault --config stacks.json

Or run everything at once:

python migrate.py

4. Open in Obsidian

1. Download Obsidian (free)
2. Open vault > select your NoteVault folder
3. Done. All your notes are there.

Configuration

Custom Stack Organization

Edit stacks.json to define how your notebooks group into stacks:

{
  "Business": [
    "Meeting Notes",
    "Client Projects",
    "Proposals"
  ],
  "Learning": [
    "Courses",
    "Certifications",
    "Book Notes"
  ],
  "Personal": [
    "Journal",
    "Recipes",
    "Travel"
  ]
}

Custom Output Path

# Output to a cloud-synced folder
python scripts/convert_to_markdown.py --output "C:\Users\you\OneDrive\NoteVault"
python scripts/convert_to_markdown.py --output ~/Dropbox/NoteVault
python scripts/convert_to_markdown.py --output ~/Google\ Drive/NoteVault

AI Export

# Export all notes as JSON (for RAG, embeddings, knowledge graphs)
python scripts/ai_export.py --vault ~/NoteVault --format json

# Export as JSONL (one note per line, for streaming/batch processing)
python scripts/ai_export.py --vault ~/NoteVault --format jsonl

# Export specific notebook or tag
python scripts/ai_export.py --vault ~/NoteVault --notebook "Meeting Notes"
python scripts/ai_export.py --vault ~/NoteVault --tag "AI"

Output Format

Every note becomes a .md file with structured frontmatter:

---
title: "Meeting with Client X"
created: "2024-03-15 14:30:00"
updated: "2024-03-16 09:00:00"
notebook: "Client Meetings"
source: "evernote"
tags: ["meeting", "client", "Q1"]
source_url: "https://..."
---

# Meeting with Client X

The actual note content in clean Markdown...

- [x] Action item completed
- [ ] Action item pending

![photo.jpg](_attachments/photo.jpg)

Folder Structure After Migration

NoteVault/
  Dashboard.md                  # Homepage with links to all stacks
  _INDEX.md                     # Master index
  _TAG_INDEX.md                 # Browse by tag
  _templates/                   # Note templates
    daily-note.md
    meeting-note.md
    project-note.md
    quick-note.md
  Inbox/                        # Quick capture destination
  Business Stack/               # Stack (group of notebooks)
    Client Projects/            # Notebook (group of notes)
      Meeting Note 1.md         # Note
      Meeting Note 2.md
      _attachments/
        screenshot.png
    Proposals/
      ...
  Personal Stack/
    Journal/
      ...
  search_notes.py               # Full-text search tool
  quick_note.py                 # Quick capture tool
  ai_export.py                  # AI data export tool

Tested Scale

Metric	Our Migration
Notes	6,453
Notebooks	293
Attachments	3,172
Data volume	2.2 GB
Download time	~25 minutes
Conversion time	~5 minutes
Total time	~30 minutes
Errors	0

Troubleshooting

Token Error: "Invalid token format"

Make sure you paste the token without surrounding quotes. The token should start with S=s not "S=s.

Windows Path Too Long

Some Evernote note titles are very long. The converter automatically truncates filenames to 100 characters. If you still get path errors, try a shorter output path (e.g., C:\NV instead of C:\Users\...\NoteVault).

Unicode/Emoji in Note Titles

The converter handles emoji and special characters in note titles. If you see encoding errors on Windows, the converter automatically switches to UTF-8 output.

OneDrive/Dropbox Lock Errors

Cloud sync services may lock newly created files. The converter gracefully skips cleanup operations that fail due to locks. This doesn't affect your notes.

Evernote Server Errors During Download

Some notes may fail with JDBCConnectionException. These are Evernote server-side issues. Re-run the sync command to retry:

python -m evernote_backup sync -d evernote_backup.db

Why Not [Alternative]?

Tool	Issue
Evernote's built-in export	Manual, one notebook at a time. 293 notebooks = 15+ hours of clicking.
Import2 / CloudHQ	Per-note pricing ($200-500 for large accounts). Still needs API access.
Notion import	Imports to another SaaS with its own lock-in.
Google Keep	No bulk import. Different use case (sticky notes vs. documents).
Copy-paste	Not serious for 6,000+ notes.

Contributing

Issues and pull requests welcome. Key areas for contribution:

• Linux/macOS testing and path handling
• Additional export formats (HTML, PDF)
• Notion/Obsidian/Logseq-specific optimizations
• Web UI for non-technical users
• Progress bars and ETA display
• Incremental sync (only convert new/changed notes)

License

MIT License. Use it, modify it, share it.

Credits

Built by Evangel (Ev) Oputa, Founder of Begine Fusion and Co-Founder of OnStack AI Labs.

Begine Fusion is a digital adoption company. We set up AI, CRM, automation, and growth marketing systems inside businesses, then make sure they actually run.

OnStack AI Labs — Calgary's first innovative, structured, collaborative skills and applied AI lab, working across a global ecosystem.

Technical implementation assisted by Claude Code (Anthropic).

---

If this tool saved you from Evernote jail, give it a star and share it with others stuck in the same situation.